cli-config.yaml.example

# Hermes Agent CLI Configuration
# Copy settings from this example into ~/.hermes/config.yaml, or use
# `hermes config set <section.key> <value>` to update the active profile.
# This file configures CLI behavior; only documented secret environment
# variables in .env take precedence over their corresponding settings.

# =============================================================================
# Database Configuration
# =============================================================================
# WAL is the normal default. Hermes automatically falls back to DELETE when
# SQLite reports that WAL is incompatible with the filesystem. Set this to
# "delete" explicitly for deployments whose backing filesystem is not WAL
# crash-safe, such as Linux containers bind-mounted through macOS virtiofs,
# NFS, or SMB. Hermes will not live-downgrade a database already open in WAL.
database:
  journal_mode: "wal"  # Supported values: "wal", "delete"
  # Optional WAL sizing pragmas (integers). Unset = SQLite defaults.
  # wal_autocheckpoint: 1000     # pages between automatic checkpoints
  # journal_size_limit: 67108864 # cap the WAL/journal file size in bytes
  #
  # Durability level for every state.db connection: OFF, NORMAL, FULL, EXTRA
  # (or 0-3). Unset leaves SQLite's default, which is baked in at compile time
  # (SQLITE_DEFAULT_WAL_SYNCHRONOUS) and therefore differs between the bundled
  # interpreter, a distro python3 and a Homebrew one. Set it if you need to
  # know which one you are running rather than infer it. On macOS this is a
  # floor, not a pin: values below FULL are refused because Darwin's fsync()
  # does not guarantee write ordering, while EXTRA is honored normally.
  # synchronous: FULL

# =============================================================================
# Runtime Limits
# =============================================================================
# Long-running Hermes server processes raise their RLIMIT_NOFILE soft limit to
# this value when the operating system permits it. The value is clamped to the
# hard limit and never lowers an already higher soft limit. Set to 0, false, or
# null to disable the adjustment. Default: 4096.
runtime:
  nofile_soft_limit: 4096

# =============================================================================
# Model Configuration
# =============================================================================
model:
  # Default model to use (can be overridden with --model flag)
  # Both "default" and "model" work as the key name here.
  default: "anthropic/claude-opus-4.6"
  
  # Inference provider selection:
  #   "auto"         - Auto-detect from credentials (default)
  #   "openrouter"   - OpenRouter (requires: OPENROUTER_API_KEY or OPENAI_API_KEY)
  #   "nous"         - Nous Portal OAuth (requires: hermes auth add nous)
  #   "nous-api"     - Nous Portal API key (requires: NOUS_API_KEY)
  #   "anthropic"    - Direct Anthropic API (requires: ANTHROPIC_API_KEY)
  #   "openai-codex" - OpenAI Codex (requires: hermes auth)
  #   "copilot"      - GitHub Copilot / GitHub Models (requires: GITHUB_TOKEN)
  #   "gemini"      - Use Google AI Studio direct (requires: GOOGLE_API_KEY or GEMINI_API_KEY)
  #   "zai"         - Use z.ai / ZhipuAI GLM models (requires: GLM_API_KEY)
  #   "kimi-coding"  - Kimi / Moonshot AI (requires: KIMI_API_KEY)
  #   "minimax"      - MiniMax global (requires: MINIMAX_API_KEY)
  #   "minimax-cn"   - MiniMax China (requires: MINIMAX_CN_API_KEY)
  #   "huggingface"  - Hugging Face Inference (requires: HF_TOKEN)
  #   "nvidia"       - NVIDIA NIM / build.nvidia.com (requires: NVIDIA_API_KEY)
  #   "xiaomi"       - Xiaomi MiMo (requires: XIAOMI_API_KEY)
  #   "arcee"        - Arcee AI Trinity models (requires: ARCEEAI_API_KEY)
  #   "ollama-cloud" - Ollama Cloud (requires: OLLAMA_API_KEY — https://ollama.com/settings)
  #   "deepinfra"    - DeepInfra (requires: DEEPINFRA_API_KEY)
  #   "kilocode"     - KiloCode gateway (requires: KILOCODE_API_KEY)
  #   "ai-gateway"   - Vercel AI Gateway (requires: AI_GATEWAY_API_KEY)
  #   "azure-foundry" - Microsoft Foundry / Azure OpenAI (API key or Entra ID)
  #   "lmstudio"     - LM Studio local server (optional: LM_API_KEY, defaults to http://127.0.0.1:1234/v1)
  #
  # Local servers (LM Studio, Ollama, vLLM, llama.cpp):
  #   "custom"       - Any other OpenAI-compatible endpoint. Set base_url below.
  #   Aliases: "ollama", "vllm", "llamacpp" all map to "custom".
  #   LM Studio is first-class and uses provider: "lmstudio".
  #   It works with both no-auth and auth-enabled server modes.
  #
  # Can also be overridden for a single invocation with the --provider flag.
  provider: "auto"
  
  # API configuration (falls back to OPENROUTER_API_KEY env var)
  # api_key: "your-key-here"  # Uncomment to set here instead of .env
  base_url: "https://openrouter.ai/api/v1"

  # Stream API responses from the provider (default: true). The agent core
  # prefers streaming for every turn — subagents included — for liveness
  # health-checking. Set false to force non-streaming requests for the whole
  # session (persists across mid-session model switches). Escape hatch for
  # self-hosted OpenAI-compatible servers whose streaming tool-call path is
  # broken (e.g. vLLM with --tool-call-parser qwen3_xml + a reasoning parser
  # can leak tool calls into plain text instead of returning tool_calls —
  # #72901). Orthogonal to display.streaming, which controls token rendering
  # only.
  # streaming: true

  # Azure Foundry keyless auth example:
  # provider: "azure-foundry"
  # base_url: "https://<resource>.openai.azure.com/openai/v1"
  # auth_mode: "entra_id"      # DefaultAzureCredential: az login, managed identity, workload identity, etc.
  # default: "gpt-4o"          # Deployment/model name
  # entra:
  #   scope: "https://ai.azure.com/.default"  # Optional; this is the default.

  # ── Token limits — two settings, easy to confuse ──────────────────────────
  #
  # context_length: TOTAL context window (input + output tokens combined).
  #   Controls when Hermes compresses history and validates requests.
  #   Leave unset — Hermes auto-detects the correct value from the provider.
  #   Set manually only when auto-detection is wrong (e.g. a local server with
  #   a custom num_ctx, or a proxy that doesn't expose /v1/models).
  #
  # context_length: 131072
  #
  # Output-token limits are provider-owned, not user configuration. Native
  # protocols requiring a limit receive an internal value from Hermes.

  # ── Custom request headers (optional) ─────────────────────────────────────
  #
  # default_headers: extra HTTP headers sent on every request to an
  #   OpenAI-compatible endpoint. User values take precedence over the
  #   provider/SDK defaults, so this is the supported way to override the
  #   OpenAI Python SDK's identifying headers (User-Agent: OpenAI/Python ...,
  #   X-Stainless-*) when a custom provider sits behind a gateway/WAF that
  #   rejects them — e.g. an upstream that returns "502 Upstream access
  #   forbidden" for the SDK default User-Agent but accepts a plain one.
  #   Applies on the OpenAI wire only (not native Anthropic / Bedrock).
  #
  # default_headers:
  #   User-Agent: "curl/8.7.1"
  #
  # extra_headers: accepted as an alias of default_headers (merged, with
  #   extra_headers winning when both are set) — matches the per-provider
  #   extra_headers key below.
  #
  # Per-provider variant: named providers / custom_providers entries accept an
  # extra_headers dict scoped to that endpoint only — for reverse proxies,
  # gateways, or custom auth (e.g. Cloudflare Access service tokens).
  # Merged onto SDK/provider defaults with the entry's values winning.
  # Header values are treated as secrets and are never logged.
  #
  # providers:
  #   my-proxy:
  #     base_url: "https://llm.internal.example.com/v1"
  #     key_env: "MY_PROXY_API_KEY"
  #     extra_headers:
  #       CF-Access-Client-Id: "xxxx.access"
  #       CF-Access-Client-Secret: "${CF_ACCESS_SECRET}"
  #       X-Client-Name: "hermes-agent"
# providers:
#   meta:
#     base_url: https://api.meta.ai/v1
#     api_key: ${MODEL_API_KEY}
#     # api_mode auto-detected as codex_responses for api.meta.ai; no need to set
#     # (the bundled meta-ai provider covers this — a named custom provider is
#     # only needed for a non-default Meta-compatible endpoint)
# providers:
#   router:
#     base_url: https://api.router.com/v1
#     api_key: ${RAMP_ROUTER_API_KEY}
#     # api_mode auto-detected as codex_responses for api.router.com — the
#     # Responses API is Router's native wire (chat/completions is only a
#     # compatibility shim). The bundled router provider covers this; a named
#     # custom provider is only needed for a non-default Router-compatible
#     # endpoint.


# Command-minted credentials (optional): key_cmd
# ------------------------------------------------------------------
# Enterprise gateways often issue SHORT-LIVED bearers (SSO/OIDC brokers, cloud
# IAM, internal auth proxies) rather than static API keys, so a value copied
# into .env via `key_env` is stale within the hour and every later request 401s.
# `key_cmd` names a command that PRINTS a token instead: Hermes runs it per
# request (cached until shortly before expiry), so long sessions keep working
# with no restart.
#
# Contract: print ONLY the token on stdout, either bare or as JSON with an
# "access_token" field ("expires_in" is honoured). Same shape as OAuth 2.0
# token endpoints, Claude Code's `apiKeyHelper`, `gcloud auth
# print-access-token`, and `aws ecr get-login-password`.
#
# Precedence: an explicit --api-key still wins; otherwise key_cmd is preferred
# over inline api_key / key_env on that entry.
#
# Applies to the main agent turn and to auxiliary tasks (title generation,
# context compression, vision, embedding) alike.
#
# Not to be confused with `secrets.command`, which is a different mechanism:
# that one runs a helper ONCE at startup to populate env vars for many secrets
# at the process level. Use it for a vault or keychain helper that hands back a
# KEY=VALUE blob. Use `key_cmd` when ONE provider needs a credential refreshed
# DURING a session, because a startup-time env var cannot be re-minted after it
# expires.
#
# providers:
#   my-gateway:
#     base_url: "https://gateway.internal.example.com/v1"
#     api_mode: chat_completions
#     key_cmd: "my-auth-cli print-token --profile prod"
#
# Worked example — an AI gateway that routes by model family, so one entry per
# wire format shares the same credential helper:
#
# providers:
#   dbx:                             # OpenAI-compatible (MLflow) route
#     base_url: "https://<workspace>.cloud.databricks.com/ai-gateway/mlflow/v1"
#     api_mode: chat_completions
#     model: databricks-claude-sonnet-4-6
#     key_cmd: "databricks auth token -p MY-PROFILE"
#   dbx-gpt:                         # OpenAI Responses route
#     base_url: "https://<workspace>.cloud.databricks.com/ai-gateway/openai/v1"
#     api_mode: codex_responses
#     model: databricks-gpt-5-5
#     key_cmd: "databricks auth token -p MY-PROFILE"
#   dbx-claude:                      # Anthropic Messages route
#     base_url: "https://<workspace>.cloud.databricks.com/ai-gateway/anthropic"
#     api_mode: anthropic_messages
#     model: databricks-claude-fable-5
#     key_cmd: "databricks auth token -p MY-PROFILE"

# Named provider overrides (optional)
# Use this for per-provider request timeouts, non-stream stale timeouts,
# and per-model exceptions.
# Applies to the primary turn client on every api_mode (OpenAI-wire, native
# Anthropic, and Anthropic-compatible providers), the fallback chain, and
# client rebuilds during credential rotation.  For OpenAI-wire chat
# completions (streaming and non-streaming) the configured value is also
# used as the per-request ``timeout=`` kwarg so it wins over the legacy
# HERMES_API_TIMEOUT env var (which still applies when no config is set).
# ``stale_timeout_seconds`` controls the non-streaming stale-call detector and
# wins over the legacy HERMES_API_CALL_STALE_TIMEOUT env var. Leaving these
# unset keeps the legacy defaults (HERMES_API_TIMEOUT=1800s,
# HERMES_API_CALL_STALE_TIMEOUT=90s, native Anthropic 900s). The
# implicit non-stream stale detector is auto-disabled for local endpoints
# and can scale upward for very large contexts.
#
# Not currently wired for AWS Bedrock (bedrock_converse + AnthropicBedrock
# SDK paths) — those use boto3 with its own timeout configuration.
#
# providers:
#   ollama-local:
#     request_timeout_seconds: 300   # Longer timeout for local cold-starts
#     stale_timeout_seconds: 900     # Explicitly re-enable stale detection on local endpoints
#   anthropic:
#     request_timeout_seconds: 30    # Fast-fail cloud requests
#     models:
#       claude-opus-4.6:
#         timeout_seconds: 600       # Longer timeout for extended-thinking Opus calls
#   openai-codex:
#     models:
#       gpt-5.4:
#         stale_timeout_seconds: 1800  # Longer non-stream stale timeout for slow large-context turns

# =============================================================================
# Unified Timeouts (operation deadlines)
# =============================================================================
# One place to override Hermes's internal operation deadlines (seconds).
# Keys are dotted paths resolved by agent/deadline.py:resolve_timeout().
# Precedence: this section > legacy HERMES_* env var (back-compat) > built-in
# default. 0 or a negative value disables the bound (unbounded); very large
# values are clamped to a platform-safe maximum automatically.
#
# Currently resolved keys (more paths migrate here over time — see issue #85125):
#
# timeouts:
#   tools:
#     concurrent_batch: 420   # Deadline for a parallel tool-call batch
#                             # (legacy env: HERMES_CONCURRENT_TOOL_TIMEOUT_S)
#     sequential_call: 420    # Deadline for one sequentially-executed tool call.
#                             # Defaults to concurrent_batch's value so the two
#                             # executor paths stay in sync; human waits
#                             # (approval prompts, clarify) never count against it.

# =============================================================================
# OpenRouter Provider Routing (only applies when using OpenRouter)
# =============================================================================
# Control how requests are routed across providers on OpenRouter.
# See: https://openrouter.ai/docs/guides/routing/provider-selection
#
# provider_routing:
#   # Sort strategy: "price" (default), "throughput", or "latency"
#   # Append :nitro to model name for a shortcut to throughput sorting.
#   sort: "throughput"
#
#   # Only allow these providers (provider slugs from OpenRouter)
#   # only: ["anthropic", "google"]
#
#   # Skip these providers entirely
#   # ignore: ["deepinfra", "fireworks"]
#
#   # Try providers in this order (overrides default load balancing)
#   # order: ["anthropic", "google", "together"]
#
#   # Require providers to support all parameters in your request
#   # require_parameters: true
#
#   # Data policy: "allow" (default) or "deny" to exclude providers that may store data
#   # data_collection: "deny"
#
#   # Per-model overrides: same keys, applied only when the agent is on that model
#   # (spelling-tolerant match; unset keys fall through to the flat values above).
#   # models:
#   #   "openai/gpt-6-astra":
#   #     only: ["openai"]
#   #   "anthropic/claude-fable-5.1":
#   #     only: ["anthropic"]

# =============================================================================
# OpenRouter Response Caching (only applies when using OpenRouter)
# =============================================================================
# Cache identical API responses at the OpenRouter edge for free instant replays.
# When enabled, identical requests (same model, messages, parameters) return
# cached responses with zero billing. Separate from Anthropic prompt caching.
# See: https://openrouter.ai/docs/guides/features/response-caching
#
# openrouter:
#   response_cache: true         # Enable response caching (default: true)
#   response_cache_ttl: 300      # Cache TTL in seconds, 1-86400 (default: 300)

# =============================================================================
# Git Worktree Isolation
# =============================================================================
# When enabled, each CLI session creates an isolated git worktree so multiple
# agents can work on the same repo concurrently without file collisions.
# Equivalent to always passing --worktree / -w on the command line.
#
# worktree: true    # Always create a worktree when in a git repo
# worktree: false   # Default — only create when -w flag is passed
#
# By default a new worktree branches from the freshly-fetched remote tip
# (the current branch's upstream, else the remote's default branch) so it
# starts current with the project instead of from the local clone's
# (possibly stale) HEAD. Set worktree_sync: false to branch from local HEAD
# instead — useful when offline or when you deliberately want the clone's
# exact current state as the base.
#
# worktree_sync: true   # Default — branch from the fetched remote tip
# worktree_sync: false  # Branch from local HEAD (offline / pinned base)

# =============================================================================
# Kanban Review Dispatch
# =============================================================================
# First-class review tasks are dispatched automatically by default. The worker
# is spawned with the bundled sdlc-review skill and can approve, request changes
# back to the original implementer, or escalate a genuine external blocker.
# Disable this only when every review is performed manually from the dashboard.
kanban:
  review_dispatch: true

# =============================================================================
# Terminal Tool Configuration
# =============================================================================
# Choose ONE of the following terminal configurations by uncommenting it.
# The terminal tool executes commands in the specified environment.

# -----------------------------------------------------------------------------
# OPTION 1: Local execution (default)
# Commands run directly on your machine in the current directory
# -----------------------------------------------------------------------------
# Working directory behavior:
#   - CLI (`hermes` command): Uses "." (current directory where you run hermes)
#   - Gateway/messaging/cron: Uses terminal.cwd here; legacy .env cwd values are deprecated
cron:
  catch_up_missed: true  # False skips past-grace recurring misses after planned downtime.

terminal:
  backend: "local"
  cwd: "."  # For local backend: "." = current directory. Ignored for remote backends unless a backend documents otherwise.
  # Desktop xterm font. Install the font locally; Nerd Fonts render Powerlevel10k glyphs.
  # font_family: "MesloLGS NF"  # Also accepts a CSS stack; blank uses bundled JetBrains Mono.
  timeout: 180
  # HOME policy for tool subprocesses:
  #   auto    - default: host uses your real HOME; containers use HERMES_HOME/home
  #   real    - force your real OS-user HOME
  #   profile - force HERMES_HOME/home for strict per-profile CLI config isolation
  home_mode: "auto"
  docker_mount_cwd_to_workspace: false  # SECURITY: off by default. Opt in to mount the launch cwd into Docker /workspace.
  lifetime_seconds: 300
  # sudo_password: "hunter2"  # Optional: pipe a sudo password via sudo -S. SECURITY WARNING: plaintext.
  # sudo_password: ""         # Explicit empty password: try empty and never open the interactive sudo prompt.

# -----------------------------------------------------------------------------
# OPTION 2: SSH remote execution
# Commands run on a remote server - agent code stays local (sandboxed)
# Great for: keeping agent isolated from its own code, using powerful remote hardware
# -----------------------------------------------------------------------------
# terminal:
#   backend: "ssh"
#   cwd: "/home/myuser/project"  # Path on the REMOTE server
#   timeout: 180
#   lifetime_seconds: 300
#   ssh_host: "my-server.example.com"
#   ssh_user: "myuser"
#   ssh_port: 22
#   ssh_key: "~/.ssh/id_rsa"  # Optional - uses ssh-agent if not specified

# -----------------------------------------------------------------------------
# OPTION 3: Docker container
# Commands run in an isolated Docker container
# Great for: reproducible environments, testing, isolation
# -----------------------------------------------------------------------------
# terminal:
#   backend: "docker"
#   cwd: "/workspace"  # Path INSIDE the container (default: /)
#   timeout: 180
#   lifetime_seconds: 300
#   docker_image: "nikolaik/python-nodejs:python3.11-nodejs20"
#   docker_mount_cwd_to_workspace: true   # Explicit opt-in: mount your launch cwd into /workspace
#   # Optional: run the container as your host user's uid:gid so files written
#   # into bind-mounted dirs are owned by you, not root. Drops SETUID/SETGID
#   # caps too since no gosu privilege drop is needed. Leave off if your
#   # chosen docker_image expects to start as root.
#   docker_run_as_host_user: true
#   # Optional: explicitly forward selected env vars into Docker.
#   # These values come from your current shell first, then ~/.hermes/.env.
#   # Warning: anything forwarded here is visible to commands run in the container.
#   docker_forward_env:
#     - "GITHUB_TOKEN"
#     - "NPM_TOKEN"
#   # Optional: extra flags passed verbatim to docker run (appended after security defaults).
#   # Useful for adding capabilities (e.g. apt installs needing SETUID) or custom options.
#   # Example: add a Linux capability not included by default
#   # docker_extra_args:
#   #   - "--cap-add"
#   #   - "SETUID"

# -----------------------------------------------------------------------------
# OPTION 4: Singularity/Apptainer container
# Commands run in a Singularity container (common in HPC environments)
# Great for: HPC clusters, shared compute environments
# -----------------------------------------------------------------------------
# terminal:
#   backend: "singularity"
#   cwd: "/workspace"  # Path INSIDE the container (default: /root)
#   timeout: 180
#   lifetime_seconds: 300
#   singularity_image: "docker://nikolaik/python-nodejs:python3.11-nodejs20"

# -----------------------------------------------------------------------------
# OPTION 5: Modal cloud execution
# Commands run on Modal's cloud infrastructure
# Great for: GPU access, scalable compute, serverless execution
# -----------------------------------------------------------------------------
# terminal:
#   backend: "modal"
#   cwd: "/workspace"  # Path INSIDE the sandbox (default: /root)
#   timeout: 180
#   lifetime_seconds: 300
#   modal_image: "nikolaik/python-nodejs:python3.11-nodejs20"

# -----------------------------------------------------------------------------
# OPTION 6: Daytona cloud execution
# Commands run in Daytona cloud sandboxes
# Great for: Cloud dev environments, persistent workspaces, team collaboration
# Requires: pip install daytona, DAYTONA_API_KEY env var
# -----------------------------------------------------------------------------
# terminal:
#   backend: "daytona"
#   cwd: "~"
#   timeout: 180
#   lifetime_seconds: 300
#   daytona_image: "nikolaik/python-nodejs:python3.11-nodejs20"
#   container_disk: 10240          # Daytona max is 10GB per sandbox

#
# --- Container resource limits (docker, singularity, modal, daytona -- ignored for local/ssh) ---
# These settings apply to all container backends. They control the resources
# allocated to the sandbox and whether its filesystem persists across sessions.
  container_cpu: 1              # CPU cores
  container_memory: 5120        # Memory in MB (5120 = 5GB)
  container_disk: 51200         # Disk in MB (51200 = 50GB)
  container_persistent: true    # Persist filesystem across sessions (false = ephemeral)

# -----------------------------------------------------------------------------
# SUDO SUPPORT (works with ALL backends above)
# -----------------------------------------------------------------------------
# Add sudo_password to any terminal config above to enable sudo commands.
# The password is piped via `sudo -S`. Works with local, ssh, docker, etc.
#
# SECURITY WARNING: Password stored in plaintext!
#
# INTERACTIVE PROMPT: If sudo_password is unset and the CLI is running,
# you'll be prompted to enter your password when sudo is needed:
# - 45-second timeout (auto-skips if no input)
# - Press Enter to skip (command fails gracefully)
# - Password is hidden while typing
# - Password is cached for the session
#
# EMPTY PASSWORDS: Setting sudo_password to an explicit empty string is different
# from leaving it unset. Hermes will try an empty password via `sudo -S` and
# will not open the interactive prompt. This is useful for passwordless sudo,
# Touch ID sudo setups, and environments where prompting is just noise.
#
# ALTERNATIVES:
# - SSH backend: Configure passwordless sudo on the remote server
# - Containers: Run as root inside the container (no sudo needed)
# - Local: Configure /etc/sudoers for specific commands
#
# Example (add to your terminal section):
#   sudo_password: "your-password-here"

# =============================================================================
# Security Scanning (tirith)
# =============================================================================
# Optional pre-exec command security scanning via tirith.
# Detects homograph URLs, pipe-to-shell, terminal injection, env manipulation.
# Install: brew install sheeki03/tap/tirith
# Docs: https://github.com/sheeki03/tirith
#
# security:
#   tirith_enabled: true        # Enable/disable tirith scanning
#   tirith_path: "tirith"       # Path to tirith binary (supports ~ expansion)
#   tirith_timeout: 5           # Scan timeout in seconds
#   tirith_fail_open: true      # Allow commands if tirith unavailable
#   approval:
#     transport: builtin        # Or an explicitly enabled plugin transport name
#     transport_fallback: deny  # Set builtin to opt into fallback on transport failure

# =============================================================================
# Browser Tool Configuration
# =============================================================================
browser:
  # Inactivity timeout in seconds - browser sessions are automatically closed
  # after this period of no activity between agent loops (default: 120 = 2 minutes)
  inactivity_timeout: 120
  # Let an authenticated browser extension register as the controller for an
  # existing Hermes session. Disabled by default. Local API registration also
  # requires the API server bearer key to be configured.
  extension_control:
    enabled: false

  # Maximum characters of snapshot content before truncate-and-store.
  # Increase for long pages (for example, documentation or financial reports),
  # or decrease to reduce context usage. Minimum: 1000; default: 15000
  # (same per-page budget as web_extract).
  # snapshot_threshold: 15000

# =============================================================================
# Tool Loop Guardrails
# =============================================================================
# Soft warnings are enabled by default. They append guidance to repeated failed
# or non-progressing tool results but still let the tool execute. Hard stops stay
# opt-in for interactive CLI/TUI/Desktop/ACP sessions, but default on for unattended
# gateway/cron sessions where nobody is present to interrupt a model that
# ignores loop warnings.
tool_loop_guardrails:
  warnings_enabled: true
  hard_stop_enabled: false
  non_interactive_hard_stop_enabled: true
  warn_after:
    exact_failure: 2
    same_tool_failure: 3
    idempotent_no_progress: 2
  hard_stop_after:
    exact_failure: 5
    same_tool_failure: 8
    idempotent_no_progress: 5

# =============================================================================
# Context Compression (Auto-shrinks long conversations)
# =============================================================================
# When conversation approaches model's context limit, middle turns are
# automatically summarized to free up space while preserving important context.
#
# HOW IT WORKS:
# 1. Tracks actual token usage from API responses (not estimates)
# 2. When prompt_tokens >= threshold% of model's context_length, triggers compression
# 3. Protects first 3 turns (system prompt, initial request, first response)
# 4. Protects last N turns (default 20 messages = ~10 full turns of recent context)
# 5. Summarizes middle turns using a fast/cheap model
# 6. Inserts summary as a user message, continues conversation seamlessly
#
# Post-compression tail budget is target_ratio × threshold × context_length:
#   200K context, threshold 0.50, ratio 0.20 → 20K tokens of recent tail preserved
#   1M   context, threshold 0.50, ratio 0.20 → 100K tokens of recent tail preserved
#
compression:
  # Enable automatic context compression (default: true)
  # Set to false if you prefer to manage context manually or want errors on overflow
  enabled: true

  # Fail closed before lossy compaction unless an active memory provider that
  # implements the pre-compress checkpoint contract (API v2) confirms its
  # durable checkpoint (default: false). With this on and no checkpoint, the
  # compaction attempt errors with BLOCKED_MISSING_PREREQUISITE and the
  # uncompressed transcript is preserved for a later retry. The gate binds to
  # every compaction authority: server-side native compaction
  # (codex_responses_native) is suppressed while armed, post-turn
  # micro-compaction is forced off (no checkpoint hook in its path), and the
  # codex_app_server API mode is refused at agent init (the codex agent
  # compacts its own thread — no checkpoint can be guaranteed there). Only
  # enable it with a checkpoint-capable provider configured — see
  # website/docs/developer-guide/memory-provider-plugin.md
  # ("Pre-Compress Checkpoints").
  checkpoint_required: false

  # Opt-in compression progress notices on chat platforms (default: false).
  # By design, routine automatic compression is SILENT on human-facing chat
  # gateways (Telegram, Discord, Slack, ...) — it happens in the background
  # with server-side logging only. Set true to also deliver the routine
  # progress statuses (compacting started, preflight/pre-API compression,
  # idle compaction, retry progress, and the compaction-complete notice) to
  # chat platforms. Unrelated operational noise (auxiliary model failures,
  # provider retry chatter) stays suppressed either way, and compression
  # FAILURE notices + manual /compress feedback are always visible
  # regardless of this setting. (#52995)
  progress_notices: false
  
  # Trigger compression at this % of model's context limit (default: 0.50 = 50%)
  # Lower values = more aggressive compression, higher values = compress later
  # Models with context windows below 512K are floored at 0.75 (raise-only) so
  # compaction doesn't fire with half the window still free; set above 0.75 to override.
  threshold: 0.50

  # Per-model threshold overrides: keys are substring-matched against the model
  # name (longest match wins). Useful when some models need different compaction
  # points — e.g. a 1M-context model can compress later (0.30) while a 128K
  # model needs to compress earlier (0.60). The small-context floor (75% for
  # <512K models) still applies on top of per-model overrides.
  # model_thresholds:
  #   "glm-5.2": 0.40
  #   "claude-sonnet": 0.35
  #   "gpt-5": 0.30

  # Optional absolute token cap for the compression trigger (default: null = disabled).
  # When set, compression fires at the LOWER of the ratio-based threshold and this
  # absolute token count — first-fires-wins. It never fires later than this count
  # regardless of which model is active (useful when switching between models with
  # very different context windows). Clamped to the model's context length at
  # apply-time, so a cap above the window is a no-op (ratio-based threshold wins).
  # Survives model switches and fallback activations.
  # threshold_tokens: 200000

  # Existing Codex gpt-5.5 behavior: raise Hermes' compaction trigger to 85%
  # for the ChatGPT Codex OAuth route. Set false to opt back down to threshold.
  codex_gpt55_autoraise: true

  # Fraction of the threshold to preserve as recent tail (default: 0.20 = 20%)
  # e.g. 20% of 50% threshold = 10% of total context kept as recent messages.
  # Summary output is separately capped at 12K tokens (Gemini output limit).
  # Range: 0.10 - 0.80
  target_ratio: 0.20

  # Number of most-recent messages to always preserve (default: 20 ≈ 10 full turns)
  # Higher values keep more recent conversation intact at the cost of more aggressive
  # compression of older turns.
  protect_last_n: 20

  # Minimum number of REAL (actionable) user messages guaranteed to survive in
  # the uncompressed tail (default: 1 = the existing single last-user anchor,
  # behavior-preserving). Raise to e.g. 3 to keep the last 3 real user turns
  # verbatim even when bulky tool outputs fill the tail token budget — blank
  # platform echoes, compaction handoffs, and synthetic continuation rows never
  # count toward N. The tail can exceed the token budget when this pulls the
  # cut back; the guarantee wins over the budget by design.
  min_tail_user_messages: 1

  # Compression retry rounds before a turn gives up with "max compression
  # attempts reached" (default: 3, same as the previous hardcoded value).
  # Raise (e.g. 6) for tool-schema-heavy sessions where 3 rounds cannot bring
  # the request estimate under the threshold. Validated >= 1, hard cap 10.
  max_attempts: 3

  # Codex app-server (codex CLI runtime) thread-compaction mode. The codex
  # agent owns the real thread context on this runtime, so Hermes' summarizer
  # cannot shrink it — compaction goes through the app server instead.
  #   native = let Codex decide when to compact its own thread (default)
  #   hermes = let Hermes threshold trigger Codex thread/compact/start
  #   off    = Hermes will not auto-trigger compaction; Codex may still compact natively
  codex_app_server_auto: native

  # Native OpenAI Responses server-side compaction (default: false). When true,
  # gpt-5.6-family models on the DIRECT OpenAI API (api.openai.com) or a ChatGPT
  # Codex subscription compact server-side: OpenAI prunes older context into an
  # encrypted checkpoint that Hermes replays on later turns. No other provider,
  # route, or model is affected. Hermes' local compression stays armed as the
  # fallback and still handles every non-eligible session.
  codex_responses_native: false

  # Optional absolute server compaction trigger in input tokens. The default
  # null value follows the resolved local compression trigger with an 8192 token
  # safety margin. For example, a local trigger of 765000 selects 756808.
  # A positive integer stays absolute and only clamps downward when needed so
  # the server compacts first. Invalid values use this automatic behavior. If
  # the local trigger is unavailable, automatic mode uses 200000.
  codex_responses_compact_threshold: null

  # Number of non-system messages to protect at the head of the transcript, in
  # ADDITION to the system prompt (which is always implicitly protected).
  # Head messages are NEVER summarized — they survive every compression
  # indefinitely. This gives stable early context for short/medium sessions,
  # but in long-running sessions that rely on rolling compaction the pinned
  # opening turns may not match how you want the session framed over time.
  # Set to 0 to preserve ONLY the system prompt (plus the rolling summary
  # and recent tail) — the cleanest configuration for long-running sessions.
  # Default 3 preserves the system prompt plus the first three non-system
  # head messages, matching the pre-feature behaviour.
  protect_first_n: 3

  # Idle compaction (default: 0 = disabled). When > 0, a session that resumes
  # after at least this many seconds of inactivity compacts its accumulated
  # history up front, before the first reply, so a long-lived thread you come
  # back to later doesn't re-read its full stale context on every turn.
  # Time-based, so it complements (does not replace) the size-based `threshold`
  # above. It is skipped when the context is already small (at or below the
  # post-compression target = threshold × target_ratio), so it never wastes a
  # summarization on a short idle thread. Example: 1800 = compact after 30 min idle.
  idle_compact_after_seconds: 0

  # Gateway session-hygiene turn-hold budget (default: 10). Max seconds an
  # arriving user turn is held while a still-streaming hygiene summary
  # finishes. Distinct from hygiene_timeout_seconds (compressor inactivity
  # budget): this bounds user-visible latency so chat transports (Telegram
  # ~30s) do not drop a silent connection. On expiry the turn proceeds
  # uncompressed; the detached worker keeps its commit admission (when the
  # commit is watermark-fenced) and the summary is adopted at the next safe
  # boundary. Thinking-model summarizers often need longer than 10s to emit
  # the first content token — raise to 300 (or >= your summarizer's real
  # time-to-first-content) only if you want THIS turn to wait for the
  # compression instead of adopting it one turn late.
  hygiene_max_turn_hold_seconds: 10

  # Proactive tool-result prune (default: 0 = disabled). Opt-in token trigger
  # for a deterministic, no-LLM prune of OLD tool-result payloads, run
  # independently of `threshold` above. On large-window models (512K/1M) the
  # ratio threshold rarely fires, so bulky tool outputs (terminal dumps, file
  # reads, web extracts) ride along in history and get re-billed every turn.
  # When re-sent history exceeds this many tokens, the prune dedupes identical
  # results, summarizes older oversized ones, and truncates large tool-call
  # arguments — protecting the most recent `protect_last_n` messages and never
  # calling the model. Try 48000 to enable. Built-in compressor engine only;
  # other context engines inherit a safe no-op.
  # NOTE: a committed prune rewrites already-sent history, which invalidates
  # the provider's prompt-cache prefix — the min_reclaim gate below keeps
  # those cache breaks episodic (like a compression boundary) instead of
  # per-turn.
  proactive_prune_tokens: 0

  # The prune's summarize pass only touches tool results larger than this many
  # characters (clamped to >= 200 so a generated summary can't be
  # re-summarized). Default 8000.
  proactive_prune_min_result_chars: 8000

  # A proactive prune only COMMITS when it reclaims at least this many tokens
  # (measured on the pruned output). This is the prompt-cache hysteresis gate:
  # one meaningful, amortized cache break per batch of stale tool output
  # instead of a tiny break on every tool iteration. 0 = commit any non-zero
  # prune. Default 4096.
  proactive_prune_min_reclaim_tokens: 4096

  # To pin a specific model/provider for compression summaries, use the
  # auxiliary section below (auxiliary.compression.provider / model).

# =============================================================================
# Tool-result budget (optional)
# =============================================================================
# Controls when a large tool result is spilled to disk (full output saved to
# $HERMES_HOME/cache/spillover, preview + path kept in context). MCP tool
# results (tools named mcp_*) spill at a tighter 50,000-char threshold than
# the generic 100K default: MCP servers routinely return un-paginated 20-50K
# payloads that bloat context and slow every subsequent turn. Nothing is
# lost — the full result is on disk and readable with read_file.
#
# tool_budget:
#   mcp_result_size_chars: 50000   # per-result spillover threshold for mcp_* tools

# =============================================================================
# Anthropic prompt caching TTL
# =============================================================================
# When prompt caching is active (Claude via OpenRouter or native Anthropic),
# Anthropic supports two TTL tiers for cached prefixes: "5m" (default) and
# "1h". Other values are ignored and "5m" is used.
#
prompt_caching:
  cache_ttl: "5m" # use "1h" for long sessions with pauses between turns

# =============================================================================
# Auxiliary Models (Advanced — Experimental)
# =============================================================================
# Hermes uses lightweight "auxiliary" models for side tasks: image analysis,
# browser screenshot analysis, web page summarization, TTS audio-tag insertion,
# and context compression.
#
# By default these use Gemini Flash via OpenRouter or Nous Portal and are
# auto-detected from your credentials.  You do NOT need to change anything
# here for normal usage.
#
# WARNING: Overriding these with providers other than OpenRouter or Nous Portal
# is EXPERIMENTAL and may not work.  Not all models/providers support vision,
# produce usable summaries, or accept the same API format.  Change at your own
# risk — if things break, reset to "auto" / empty values.
#
# Each task has its own provider + model pair so you can mix providers.
# For example: OpenRouter for vision (needs multimodal), but your main
# local endpoint for compression (just needs text).
#
# Provider options:
#   "auto"       - Best available: OpenRouter → Nous Portal → main endpoint (default)
#   "openrouter" - Force OpenRouter (requires OPENROUTER_API_KEY)
#   "nous"       - Force Nous Portal (requires: hermes auth add nous)
#   "gemini"      - Force Google AI Studio direct (requires: GOOGLE_API_KEY or GEMINI_API_KEY)
#   "ollama-cloud" - Ollama Cloud (requires: OLLAMA_API_KEY)
#   "codex"       - Force Codex OAuth (requires: hermes model → Codex).
#                  Uses gpt-5.3-codex which supports vision.
#   "main"       - Use your custom endpoint (OPENAI_BASE_URL + OPENAI_API_KEY).
#                  Works with OpenAI API, local models, or any OpenAI-compatible
#                  endpoint.  Also falls back to Codex OAuth and API-key providers.
#
# Model: leave empty to use the provider's default.  When empty, OpenRouter
# uses "google/gemini-3-flash-preview" and Nous uses "gemini-3-flash".
# Other providers pick a sensible default automatically.
#
# auxiliary:
#   # Image analysis: vision_analyze tool + browser screenshots
#   vision:
#     provider: "auto"
#     model: ""              # e.g. "google/gemini-2.5-flash", "openai/gpt-4o"
#     timeout: 30            # LLM API call timeout (seconds)
#     download_timeout: 30   # Image HTTP download timeout (seconds)
#                            # Increase for slow connections or self-hosted image servers
#     reasoning_effort: ""   # Per-task thinking level: none, minimal, low, medium,
#                            # high, xhigh, max, ultra. Empty = provider default.
#                            # Works on every auxiliary task block (vision,
#                            # web_extract, compression, title_generation, curator,
#                            # background_review, moa_reference, ...). Example: run
#                            # compression at "low" and vision at "none" to cut
#                            # side-task latency/cost on reasoning models.
#
#   # Web page scraping / summarization + browser page text extraction
#   web_extract:
#     provider: "auto"
#     model: ""
#     reasoning_effort: "low"
#
#   # Gemini 3.1 TTS hidden audio-tag insertion
#   tts_audio_tags:
#     provider: "auto"       # empty model = your main chat model
#     model: ""
#     timeout: 30
#
#   # Automatic session title generation after the first exchange
#   title_generation:
#     enabled: true          # set false to disable auto-title generation
#     provider: "auto"
#     model: ""
#     timeout: 30
#     language: ""           # empty = match the user's language; or e.g. "English"
#
#   # Session search — summarizes matching past sessions
#   session_search:
#     provider: "auto"
#     model: ""
#     timeout: 30
#     max_concurrency: 3    # Limit parallel summaries to reduce request-burst 429s
#     extra_body: {}        # Provider-specific OpenAI-compatible request fields
#                           # Example for providers that support request-body
#                           # reasoning controls:
#                           # extra_body:
#                           #   enable_thinking: false
#                           # Some vLLM/Qwen deployments expect this nested:
#                           # extra_body:
#                           #   chat_template_kwargs:
#                           #     enable_thinking: false
#
#   # Auto-generated short session titles after the first exchange.
#   # Each active Discord/Telegram channel can spawn a background title
#   # call.  Cap concurrency to keep retries during provider incidents
#   # from amplifying the request burst.  Leave unset for legacy behavior
#   # (unlimited).
#   title_generation:
#     provider: "auto"
#     model: ""
#     # max_concurrency: 2    # Optional: cap simultaneous title calls
#
#   # Context compression — summarizes long sessions to shrink the prompt.
#   # Heavy and often hits the slowest provider chain.  Setting a small
#   # cap prevents many sessions from compressing simultaneously during
#   # provider degradation.  Leave unset for legacy behavior (unlimited).
#   compression:
#     provider: "auto"
#     model: ""
#     # max_concurrency: 2    # Optional: cap simultaneous compression calls
#
#   # Post-turn memory/skill self-improvement review fork. Runs after a turn
#   # when the nudge intervals fire; writes skills/memories in a daemon thread.
#   # Usage is recorded under session_model_usage task='background_review'.
#   background_review:
#     enabled: true            # false = skip automatic forks (/refine still works)
#     provider: "auto"         # or pin a cheaper model (see memory.md)
#     model: ""

# =============================================================================
# Persistent Memory
# =============================================================================
# Bounded curated memory injected into the system prompt every session.
# Two stores: MEMORY.md (agent's notes) and USER.md (user profile).
# Character limits keep the memory small and focused. The agent manages
# pruning -- when at the limit, it must consolidate or replace entries.
# Disabled by default in batch_runner.
#
memory:
  # Agent's personal notes: environment facts, conventions, things learned
  memory_enabled: true
  
  # User profile: preferences, communication style, expectations
  user_profile_enabled: true
  
  # Character limits (~2.75 chars per token, model-independent)
  memory_char_limit: 2200   # ~800 tokens
  user_char_limit: 1375     # ~500 tokens

  # Periodic memory nudge: remind the agent to consider saving memories
  # every N user turns. Set to 0 to disable. Only active when memory is enabled.
  nudge_interval: 10        # Nudge every 10 user turns (0 = disabled)

# =============================================================================
# Session Continuity (Messaging Platforms)
# =============================================================================
# Conversations persist across inactivity and daily boundaries. Use /new or
# /reset explicitly; context compression manages long conversations.

# Maximum number of simultaneously active chat sessions across CLI, TUI,
# dashboard chat, and messaging gateway. Set to null, 0, or omit to allow
# unlimited concurrent sessions. When the limit is reached, new sessions get a
# clean error while existing active sessions keep their normal behavior. This
# top-level key takes precedence over gateway.max_concurrent_sessions. The cap
# is a best-effort single-host/profile runtime guard; Hermes fails open if the
# local runtime lease registry cannot be read or locked.
max_concurrent_sessions: null

# When true, group/channel chats use one session per participant when the platform
# provides a user ID. This is the secure default and prevents users in the same
# room from sharing context, interrupts, and token costs. Set false only if you
# explicitly want one shared "room brain" per group/channel.
group_sessions_per_user: true

# =============================================================================
# Session Storage Retention (state.db)
# =============================================================================
# ~/.hermes/state.db keeps every session, message, and tool call, plus the
# FTS5 search indexes. Since #54189, auto-pruning is ON by default so the file
# stays bounded: at CLI/gateway/cron startup (at most once per
# min_interval_hours) Hermes deletes ENDED sessions whose last activity is
# older than retention_days. Open, pinned, and in-progress sessions are never
# deleted. Stale automation sessions (cron/kanban/subagent/one-shot CLI) whose
# process died without closing them are first *closed*, then aged through a
# further full retention window before removal.
#
# After a prune that removed rows, VACUUM reclaims disk space only when both
# the time throttle (min_vacuum_interval_days) has elapsed AND more than 25%
# of the file's pages are reclaimable — a dense database never pays for a full
# rewrite to reclaim a few MB.
#
# Uncomment to change the defaults shown; set auto_prune: false to keep every
# ended session forever (the pre-#54189 behavior).
# sessions:
#   auto_prune: true
#   retention_days: 90
#   vacuum_after_prune: true
#   min_vacuum_interval_days: 30
#   min_interval_hours: 24

# Optional direct endpoint for autonomous Bot Mode rooms spanning gateways.
# Leave unset for the safe default: Desktop coordinates cross-gateway rooms and
# same-gateway rooms can still continue on their own. Set this only to the
# public HTTPS base URL that another trusted Hermes gateway can reach. The API
# server key authorizes the initial invitation; scoped grants are signed by a
# separate installation-private secret and survive API-key rotation. Disbanding
# a Group Chat revokes its known peer routes; API-key rotation alone does not.
# Network access should still be limited to the peer gateways (for example with
# your VPN or firewall).
# Plain HTTP is accepted only for loopback testing; every reachable peer URL
# must use HTTPS.
#
# gateway:
#   room_link_url: "https://hermes.example.net"
#
# The equivalent environment override is HERMES_ROOM_LINK_URL. After changing
# it, restart the gateway and reopen New Group Chat > Advanced; Hermes will
# re-check the selected gateways automatically.

# Startup sweep of session rows orphaned by a dead gateway process.
# The normal disconnect cleanup runs on an in-process grace timer, so a
# gateway restart (update, crash, systemd) leaves those rows permanently
# "active". On every gateway boot — stdio TUI *and* the desktop/dashboard
# WS sidecar — tui/desktop/subagent rows whose start time AND newest
# message are both older than the session TTL (HERMES_TUI_SESSION_TTL_S,
# default 6h) are closed with end_reason "startup_orphan_reap".
# Messaging-platform sessions (Telegram, Discord, ...) are never touched;
# live in-memory sessions are excluded; swept sessions stay resumable.
#
# dashboard:
#   startup_orphan_sweep: true

# ─────────────────────────────────────────────────────────────────────────────
# API Server — per-client model routing
# ─────────────────────────────────────────────────────────────────────────────
# Route different API clients to different models/providers on a single
# Hermes deployment.  Clients choose a backend by sending a specific string
# as the OpenAI ``model`` field.  Unmapped model values fall back to the
# global model configured in the ``model:`` section above, and an explicit
# session /model override always wins over a route.
#
# Configure via the ``platforms.api_server.extra.model_routes`` gateway
# config block:
#
#   platforms:
#     api_server:
#       enabled: true
#       extra:
#         key: "your-api-server-secret"
#         model_routes:
#           # Xiaozhi clients send model="minimax-m2" → routed to MiniMax via OpenRouter
#           minimax-m2:
#             model: "minimax/minimax-m1"
#             provider: "openrouter"          # optional — overrides global provider
#             # api_key: "sk-..."             # optional — per-route UPSTREAM provider
#             #                                 key (NOT caller auth; never logged)
#             # base_url: "https://..."       # optional — per-route base URL
#           # GPT clients keep their own alias
#           gpt-5:
#             model: "openai/gpt-5"
#             provider: "openrouter"
#
# Configured aliases are automatically listed by GET /v1/models so clients
# can discover them without manual coordination.  Caller authentication is
# unchanged: every request still authenticates with the global API server
# key (``extra.key`` / API_SERVER_KEY).

# ─────────────────────────────────────────────────────────────────────────────
# Gateway Streaming
# ─────────────────────────────────────────────────────────────────────────────
# Stream tokens to messaging platforms in real-time. The bot sends a message
# on first token, then progressively edits it as more tokens arrive.
# Disabled by default — enable to try the streaming UX on Telegram/Discord/Slack.
# For Telegram, partial edits are sent as plain text and only the final edit uses MarkdownV2.
streaming:
  enabled: false
  # transport: edit           # "edit" = progressive editMessageText
  # edit_interval: 0.3        # seconds between message edits
  # buffer_threshold: 40      # chars before forcing an edit flush
  # cursor: " ▉"              # cursor shown during streaming

# =============================================================================
# Skills Configuration
# =============================================================================
# Skills are reusable procedures the agent can load and follow. The agent can
# also create new skills after completing complex tasks.
#
skills:
  # Nudge the agent to create skills after complex tasks.
  # Every N tool-calling iterations, remind the model to consider saving a skill.
  # Set to 0 to disable.
  creation_nudge_interval: 15

  # External skill directories — share skills across tools/agents without
  # copying them into ~/.hermes/skills/.  Each path is expanded (~ and ${VAR})
  # and resolved to an absolute path.  External dirs are read-only: skill
  # creation always writes to ~/.hermes/skills/.  Local skills take precedence
  # when names collide.
  # external_dirs:
  #   - ~/.agents/skills
  #   - /home/shared/team-skills

# =============================================================================
# Agent Behavior
# =============================================================================
agent:
  # Maximum tool-calling iterations per conversation (default: 500)
  # Higher = more room for complex tasks, but costs more tokens
  # Recommended: 20-30 for focused tasks, 50-100 for open exploration
  max_turns: 500

  # Inactivity timeout for gateway agent runs (seconds, 0 = unlimited).
  # The agent can run indefinitely when actively calling tools or receiving
  # API responses.  Only fires after the agent has been idle for this duration.
  # gateway_timeout: 1800

  # Maximum time an alias routing key waits for an active turn holding the same
  # resolved session lease. On expiry Hermes rejects this inbound message and
  # asks the user to resend rather than running it without serialization.
  # Non-positive values fall back to the 5-second default.
  # gateway_turn_lease_timeout: 5

  # Staged warning: send a warning before escalating to full timeout.
  # Fires once per run when inactivity reaches this threshold (seconds).
  # Set to 0 to disable the warning.
  # gateway_timeout_warning: 900

  # Session stall watchdog (seconds). When a busy session has a pending
  # inbound follow-up and the agent activity clock is idle this long, the
  # gateway logs a WARNING and notifies the user to try /new. Does not kill
  # the turn (see gateway_timeout). 0 = disable. Default 300.
  # session_stall_timeout: 300

  # Transcript-sanitiser repeated-heal escalation (#96870). The pre-send
  # sanitiser silently repairs empty non-final turns on the wire copy; after
  # this many heal passes within a 10-minute session window Hermes logs one
  # ERROR (with session id + heal pattern) and sends a ONE-TIME out-of-band
  # notice to the user suggesting /debug share or `hermes doctor`. The notice
  # goes through the status channel only — conversation context and prompt
  # caching are untouched. Set 0 to disable escalation (per-window WARNINGs
  # still fire). Default 3.
  # sanitizer_heal_escalation_threshold: 3

  # Related in-agent compression timeouts (they live under the top-level
  # compression: block, shown here for discoverability next to the stall
  # watchdog they complement — a hung compression is a common stall cause):
  # compression:
  #   context_timeout_seconds: 120          # inactivity budget for in-agent
  #                                         # compress_context (0 = disable)
  #   context_total_ceiling_seconds: 600    # absolute cap on the pre-commit
  #                                         # wait even while tokens stream

  # Graceful drain timeout for gateway stop/restart (seconds).
  # Default 0 = no drain: a restart interrupts in-flight agents immediately,
  # cleans up, and exits. Set a positive value only if you want a grace
  # window on /restart, and keep it well under systemd's TimeoutStopSec.
  # restart_drain_timeout: 0

  # Cron-only floor under the same drain (seconds). Default 30.
  # restart_drain_timeout above is written for chat turns, which are cheap to
  # interrupt: the user is told the gateway is restarting and the session
  # resumes on their next message. A cron run has no such safety net — it is
  # recorded in jobs.json as a permanent failure, nobody is waiting on it, and
  # a recurring job simply skips to its next schedule. So in-flight cron work
  # gets its own grace window instead of inheriting the 0 above.
  # Clamped at runtime to the shutdown-watchdog leash (restart_drain_timeout
  # + 60s) minus teardown headroom, so values past ~50s need a matching
  # TimeoutStopSec bump to take effect. Set 0 to opt out and drain cron on
  # restart_drain_timeout like before.
  # cron_drain_timeout: 30

  # Upper bound (seconds) a submitted prompt waits for the deferred agent
  # build (MCP discovery, model metadata, skills scan) before failing with a
  # visible error. The wait is patient — the message is delivered as soon as
  # the build completes, and a progress notice is shown past 30s — so this cap
  # only fires on a genuinely hung build. Raise it for deployments with many
  # slow or unreachable MCP servers. Default 600.
  # build_wait_timeout: 600

  # Max app-level retry attempts for API errors (connection drops, provider
  # timeouts, 5xx, etc.) before the agent surfaces the failure. Lower this
  # to 1 if you use fallback providers and want fast failover on flaky
  # primaries (default 3). The OpenAI SDK does its own low-level retries
  # underneath this wrapper — this is the Hermes-level loop.
  # api_max_retries: 3

  # After the agent edits code without fresh passing verification, nudge it to
  # verify before finishing. The default "auto" enables it on interactive
  # coding surfaces (CLI, TUI, desktop) and programmatic callers, and disables
  # it on conversational messaging surfaces (Telegram, Discord, etc.) where the
  # verification summary would reach a human as chat noise. Set true or false to
  # force it on or off; the HERMES_VERIFY_ON_STOP env var (1/0) takes precedence.
  # verify_on_stop: auto

  # Standing operator instructions for the coding posture (when Hermes is in a
  # code workspace). Appended to the coding brief as an extra system block, so
  # you can pin project-wide workflow rules without editing the shipped brief.
  # Accepts a string or a list of strings. Takes effect next session.
  # coding_instructions:
  #   - "For UI work, don't run tsc/lint until I approve the look."
  #   - "Clean the diff before you commit and push."

  # When verify-on-stop finds edited code without fresh verification evidence,
  # append guidance for creative UI work (avoid broad tsc/lint/test before visual
  # approval) and clean-diff expectations. Set false to keep that nudge terse.
  # verify_guidance: true

  # A `pre_verify` hook (plugin or shell, see Event Hooks docs) can keep the
  # agent going one more turn to verify/clean before finishing. This caps how
  # many times one turn may be nudged to continue, so a hook can't trap the loop.
  # Default 3.
  # max_verify_nudges: 3

  # Enable verbose logging
  verbose: false
  
  # Reasoning effort level (OpenRouter and Nous Portal)
  # Controls how much "thinking" the model does before responding.
  # Options: "xhigh" (max), "high", "medium", "low", "minimal", "none" (disable)
  reasoning_effort: "medium"
  
  # Per-model reasoning effort overrides (optional dict)
  # Key: any sensible model spelling works (exact, dots↔dashes interchangeable,
  #      provider prefix optional). First match wins.
  # Value: reasoning effort level (same options as reasoning_effort)
  # Override the global reasoning_effort for that specific model.
  # NOTE: no `hermes config set` support for this key -- edit YAML directly.
  # reasoning_overrides:
  #   "openrouter/anthropic/claude-opus-4.5": "xhigh"
  #   "openai/gpt-5": "low"
  #   "claude-opus-4.6": "high"           # bare model name also works
  #   "deepseek/deepseek-v4-pro": "xhigh" # dots and dashes are interchangeable
  reasoning_overrides: {}

  # Fast mode (OpenAI Priority Processing / xAI Grok 4.6 / Anthropic Fast Mode
  # on Opus 4.8+). Premium pricing; only sent to first-party endpoints.
  #   "" / "normal" - off (default)
  #   "fast"        - every request
  #   "auto"        - only the first fast_auto_seconds of every turn
  #   "cold"        - that window on the first turn of a session only
  # Also: /fast normal|fast|auto|cold [--global]
  service_tier: ""
  fast_auto_seconds: 60

  # Custom personalities (use with /personality command).
  # Built-ins (helpful, concise, technical, creative, teacher, kawaii, catgirl,
  # pirate, shakespeare, surfer, noir, uwu, philosopher, hype) are always
  # available on every surface — defined once in hermes_cli/personality.py.
  # Entries here ADD new personalities or OVERRIDE a built-in by name.
  # The active selection is stored in display.personality (never here, and
  # never in agent.system_prompt — that field is your manual system prompt).
  personalities: {}
  #   mentor: "You are a supportive mentor. Guide, don't lecture."
  #   reviewer:
  #     system_prompt: "You are a meticulous code reviewer."
  #     tone: "direct"
  #     style: "terse"

# =============================================================================
# Gateway Lifecycle
# =============================================================================
gateway:
  # Post-interrupt grace for running agents during an unexpected SIGTERM.
  # Adapter, bridge, and database teardown starts after this many seconds even
  # if an agent has not unwound. Keep it below the service-manager stop budget.
  # signal_interrupt_grace_timeout: 1

  # Let platform adapters honor HTTP_PROXY / HTTPS_PROXY / NO_PROXY (and
  # SSL_CERT_FILE) from the process environment, plus macOS system-proxy
  # auto-detection. Set to false when the gateway inherits a proxy it must not
  # use (e.g. a Windows Scheduled Task picking up a local Clash/V2Ray proxy that
  # isn't running). Explicit per-platform vars like DISCORD_PROXY still apply.
  # trust_env: true

# =============================================================================
# Toolsets
# =============================================================================
# Control which tools the agent has access to.
# Use `hermes tools` to interactively enable/disable tools per platform.

# =============================================================================
# Platform Toolsets (per-platform tool configuration)
# =============================================================================
# Override which toolsets are available on each platform.
# If a platform isn't listed here, its built-in default is used.
#
# You can use EITHER:
#   - A preset like "hermes-cli" or "hermes-telegram" (curated tool set)
#   - A list of individual toolsets to compose your own (see list below)
#
# Supported platform keys: cli, telegram, discord, whatsapp, slack, qqbot, teams, google_chat
#
# Examples:
#
#   # Use presets (same as defaults):
#   platform_toolsets:
#     cli: [hermes-cli]
#     telegram: [hermes-telegram]
#
#   # Custom: give Telegram only web + terminal + file + planning:
#   platform_toolsets:
#     telegram: [web, terminal, file, todo]
#
#   # Custom: CLI without browser or image gen:
#   platform_toolsets:
#     cli: [web, terminal, file, skills, todo, tts, cronjob]
#
#   # Restrictive: Discord gets read-only tools only:
#   platform_toolsets:
#     discord: [web, vision, skills, todo]
#
# If not set, defaults are:
#   cli:           hermes-cli            (everything + cronjob management)
#   telegram:      hermes-telegram       (terminal, file, web, vision, image, tts, browser, skills, todo, cronjob, messaging)
#   discord:       hermes-discord        (same as telegram)
#   whatsapp:      hermes-whatsapp       (same as telegram)
#   slack:         hermes-slack          (same as telegram)
#   signal:        hermes-signal         (same as telegram)
#   homeassistant: hermes-homeassistant  (same as telegram)
#   qqbot:            hermes-qqbot            (same as telegram)
#   teams:            hermes-teams            (same as telegram)
#   google_chat:      hermes-google_chat      (same as telegram)
#
platform_toolsets:
  cli: [hermes-cli]
  telegram: [hermes-telegram]
  discord: [hermes-discord]
  whatsapp: [hermes-whatsapp]
  slack: [hermes-slack]
  signal: [hermes-signal]
  homeassistant: [hermes-homeassistant]
  qqbot: [hermes-qqbot]
  yuanbao: [hermes-yuanbao]
  teams: [hermes-teams]
  google_chat: [hermes-google_chat]

# =============================================================================
# Gateway Platform Settings
# =============================================================================
# Optional per-platform messaging settings.
# Platform-specific knobs live under `extra`.
#
# platforms:
#   telegram:
#     reply_to_mode: "first"  # off | first | all
#     # guest_mode lets explicit @mentions from non-allowlisted groups through.
#     # Default false; ordinary messages, replies, and regex wake words stay blocked.
#     guest_mode: false
#     # allowed_chats: ["-1001234567890"]
#     extra:
#       disable_link_previews: false  # Set true to suppress Telegram URL previews in bot messages
#       rich_messages: false          # Bot API 10.1 rich messages (tables/task lists/details/math); default false for copyable legacy MarkdownV2, set true to opt in
#       rich_drafts: false            # Experimental rich draft previews during Telegram DM streaming; default false because Telegram Desktop/macOS can visually overlay draft frames
#       command_menu:
#         # Telegram allows up to 100 BotCommands; Hermes defaults to 60 so
#         # all built-in commands plus common skill commands stay visible
#         # while remaining under Telegram's payload-size limit. Clamped 1..100.
#         max_commands: 60
#         # prepend = user priority first, then Hermes defaults
#         # append  = Hermes defaults first, then user priority
#         # replace = only the list below defines priority
#         priority_mode: prepend
#         # Priority is applied across core + plugin + skill commands before
#         # the cap, so a listed skill command always keeps a menu slot.
#         priority:
#           - my_plugin_command
#           - my-important-skill
#   slack:
#     extra:
#       # Render live tool calls as Slack-native plan/task cards. This explicit
#       # opt-in works even though Slack text tool_progress defaults to off.
#       native_task_cards: false
#       # Slack user IDs whose Web-API posts (user token, e.g. your own
#       # dashboard/mobile front-end) count as human instead of being dropped
#       # as app traffic. Narrower than allow_bots: all. Users only — never apps.
#       api_human_users: ["U0AAAAAAA", "U0BBBBBBB"]
#       # Suppress automatic link-preview cards without removing clickable links.
#       # Omit either key to preserve Slack's default for that preview type.
#       unfurl_links: false
#       unfurl_media: false
#   webhook:
#     extra:
#       # Route scripts default to a 30 second timeout. Scripts must live under
#       # the active profile's scripts directory and receive webhook JSON on stdin.
#       script_timeout_seconds: 30
#
# Discord-specific settings (config.yaml top-level, not under platforms:):
#
# discord:
#   require_mention: true            # Require @mention in server channels (default: true)
#   auto_thread: true                # Auto-create thread on @mention (default: true)
#   free_response_channels: ""       # Channel IDs where no mention is needed
#   reactions: true                  # Show processing reactions (default: true)
#   history_backfill: true           # Recover missed channel messages on mention (default: true)
#   history_backfill_limit: 50       # Max messages to scan backwards (default: 50)

# ─────────────────────────────────────────────────────────────────────────────
# Available toolsets (use these names in platform_toolsets or the toolsets list)
#
# Run `hermes chat --list-toolsets` to see all toolsets and their tools.
# Run `hermes chat --list-tools` to see every individual tool with descriptions.
# ─────────────────────────────────────────────────────────────────────────────
#
# INDIVIDUAL TOOLSETS (compose your own):
#   web          - web_search, web_extract
#   search       - web_search only (no scraping)
#   terminal     - terminal, process
#   file         - read_file, write_file, patch, search
#   browser      - browser_navigate, browser_snapshot, browser_click, browser_type,
#                  browser_scroll, browser_back, browser_press,
#                  browser_get_images, browser_vision  (requires BROWSERBASE_API_KEY)
#   vision       - vision_analyze  (requires OPENROUTER_API_KEY)
#   image_gen    - image_generate  (requires FAL_KEY)
#   skills       - skills_list, skill_view
#   skills_hub   - skill_hub (search/install/manage from online registries — user-driven only)
#   todo         - todo (in-memory task planning, no deps)
#   tts          - text_to_speech  (Edge TTS free, or ELEVENLABS/OPENAI/MINIMAX/MISTRAL key)
#   cronjob      - cronjob (create/list/update/pause/resume/run/remove scheduled tasks)
#
# PRESETS (curated bundles):
#   hermes-cli       - All of the above except send_message
#   hermes-telegram  - terminal, file, web, vision, image_gen, tts, browser,
#                      skills, todo, cronjob, send_message
#   hermes-discord   - Same as hermes-telegram
#   hermes-whatsapp  - Same as hermes-telegram
#   hermes-slack     - Same as hermes-telegram
#
# COMPOSITE:
#   debugging    - terminal + web + file
#   safe         - web + vision (no terminal access)
#   all          - Everything available
#
#   web          - Web search and content extraction (web_search, web_extract)
#   search       - Web search only, no scraping (web_search)
#   terminal     - Command execution and process management (terminal, process)
#   file         - File operations: read, write, patch, search
#   browser      - Full browser automation (navigate, click, type, screenshot, etc.)
#   vision       - Image analysis (vision_analyze)
#   image_gen    - Image generation with FLUX (image_generate)
#   skills       - Load skill documents (skills_list, skill_view)
#   todo         - Task planning and tracking for multi-step work
#   memory       - Persistent memory across sessions (personal notes + user profile)
#   session_search - Search and recall past conversations (FTS5 + Gemini Flash summarization)
#   tts          - Text-to-speech (Edge TTS free, ElevenLabs, OpenAI, MiniMax, Mistral)
#   cronjob      - Schedule and manage automated tasks (CLI-only)
#
# Composite toolsets:
#   debugging    - terminal + web + file (for troubleshooting)
#   safe         - web + vision (no terminal access)

# NOTE: The top-level "toolsets" key is deprecated and ignored.
# Tool configuration is managed per-platform via platform_toolsets above.
# Use `hermes tools` to configure interactively, or edit platform_toolsets directly.
#
# CLI override: hermes chat --toolsets terminal,web,file

# =============================================================================
# MCP (Model Context Protocol) Servers
# =============================================================================
# Connect to external MCP servers to add tools from the MCP ecosystem.
# Each server's tools are automatically discovered and registered.
# See website/docs/user-guide/features/mcp.md for full documentation.
#
# Stdio servers (spawn a subprocess):
#   command: the executable to run
#   args: command-line arguments
#   env: environment variables (only these + safe defaults passed to subprocess)
#
# HTTP servers (connect to a URL):
#   url: the MCP server endpoint
#   headers: HTTP headers (e.g., for authentication)
#
# Optional per-server settings:
#   timeout: tool call timeout in seconds (default: 120)
#   connect_timeout: initial connection timeout (default: 60)
#   keepalive_interval: liveness ping cadence in seconds (default: 180).
#     Lower it below the server's session TTL for servers that expire idle
#     sessions quickly (e.g. Unreal Engine editor MCP, ~15s), otherwise idle
#     tool calls hit an expired session and pay a slow reconnect. Floored at 5s.
#
# mcp_servers:
#   time:
#     command: uvx
#     args: ["mcp-server-time"]
#   filesystem:
#     command: npx
#     args: ["-y", "@modelcontextprotocol/server-filesystem", "/home/user"]
#   notion:
#     url: https://mcp.notion.com/mcp
#   github:
#     command: npx
#     args: ["-y", "@modelcontextprotocol/server-github"]
#     env:
#       GITHUB_PERSONAL_ACCESS_TOKEN: "ghp_..."
#
# Sampling (server-initiated LLM requests) — enabled by default.
# Per-server config under the 'sampling' key:
#   analysis:
#     command: npx
#     args: ["-y", "analysis-server"]
#     sampling:
#       enabled: true           # default: true
#       model: "gemini-3-flash" # override model (optional)
#       max_tokens_cap: 4096    # max tokens per request
#       timeout: 30             # LLM call timeout (seconds)
#       max_rpm: 10             # max requests per minute
#       allowed_models: []      # model whitelist (empty = all)
#       max_tool_rounds: 5      # tool loop limit (0 = disable)
#       log_level: "info"       # audit verbosity

# =============================================================================
# Text-to-Speech
# =============================================================================
# TTS defaults to Edge TTS unless changed in ~/.hermes/config.yaml.
# Gemini TTS supports persona/director prompt files, and Gemini 3.1 Flash TTS
# can use a hidden auxiliary rewrite pass to insert expressive square-bracket
# audio tags into the TTS script without showing tags in chat.
#
# tts:
#   provider: "gemini"
#   speed: 1.0              # global speed multiplier (provider-specific overrides this)
#   gemini:
#     model: "gemini-3.1-flash-tts-preview"
#     voice: "Kore"
#     audio_tags: false
#     persona_prompt_file: ""  # e.g. ~/.hermes/tts/radio-host.md
#   xai:
#     voice_id: "eve"       # built-in or custom voice ID from xAI Console
#     language: "en"        # BCP-47 code ("en", "pt-BR") or "auto"
#     speed: 1.0            # 0.7-1.5 playback speed
#     auto_speech_tags: false  # insert expressive audio tags via LLM rewrite
#     text_normalization: false  # normalize numbers/abbreviations/symbols
#     optimize_streaming_latency: 0  # 0-2, trades quality for lower latency
#     sample_rate: 24000    # 22050 / 24000 / 44100 / 48000
#     bit_rate: 128000      # MP3 bitrate (codec=mp3 only)

# =============================================================================
# Voice Transcription (Speech-to-Text)
# =============================================================================
# Automatically transcribe voice messages on messaging platforms.
# Providers: local (free, faster-whisper) | groq (free tier) | openai (Whisper API) | mistral (Voxtral Transcribe)
# Set the corresponding API key in .env: GROQ_API_KEY, OPENAI_API_KEY, or MISTRAL_API_KEY.
stt:
  enabled: true
  # provider: "local"          # auto-detected if omitted
  # --- Cloud pre-upload silence trim (groq/openai/mistral/xai/elevenlabs/deepinfra) ---
  # Local whisper gets Silero VAD; cloud endpoints otherwise receive raw audio —
  # silence inflates upload time, per-audio-minute billing, and hallucination risk.
  # Collapses pauses with ffmpeg client-side; clips under 12s skip the trim, and
  # on any failure the original uploads untouched.
  # cloud_trim_silence: true   # set false to always upload the original audio
  # cloud_trim_threshold_db: -40  # audio quieter than this counts as silence
  # cloud_trim_keep_ms: 300    # how much of each pause survives (keeps natural pacing)
  # Optional static transcription prompt: vocabulary/context hints for
  # prompt-capable backends. Composition is deterministic: this config value
  # is the base, then `pre_transcription` hooks run in registration order with
  # last-writer-wins semantics per field.
  #
  # Privacy: the final prompt is sent to the configured STT provider alongside
  # the audio. Do not include secrets or session context you would not send to
  # that provider.
  #
  # Length / truncation contract: Whisper-family backends (local, openai,
  # groq, deepinfra) only use the final ~224 prompt tokens, so Hermes
  # truncates longer prompts client-side (keeping the tail) with a WARNING
  # log — never an error. Other backends own their own validation.
  #
  # Provider behavior:
  #   local (faster-whisper)  -> `initial_prompt`, forwarded unchanged
  #   openai / groq          -> `prompt`, forwarded unchanged
  #   mistral                -> `prompt`, forwarded unchanged
  #   deepinfra              -> OpenAI-compatible `prompt`, unchanged
  #   local_command/xai/
  #   elevenlabs             -> unsupported; DEBUG log, then continue
  #   plugin providers       -> `prompt` in `**extra`; plugin owns limits
  # prompt: "Hermes, Teknium, Nous Research, kanban"
  local:
    model: "base"              # tiny | base | small | medium | large-v3 | turbo
    # language: ""             # auto-detect; set to "en", "es", "fr", etc. to force
    # initial_prompt: ""       # Optional faster-whisper prompt, e.g. bias Chinese output to simplified Chinese
    # --- Anti-hallucination hardening (whisper decodes junk from silence without these) ---
    # vad: true                # Silero VAD filter (default on) — silence never reaches whisper.
    #                          # Set false to restore raw behavior (e.g. transcribing music/ambient audio).
    # vad_min_silence_ms: 500  # min silence (ms) that splits speech chunks when vad is on
    # no_speech_prob_threshold: 0.6  # drop a segment only if no_speech_prob > this...
    # logprob_threshold: -1.0       # ...AND avg_logprob < this (both must hit — quiet real speech survives)
    # unload_after_idle_seconds: 0  # 0=never unload (default); e.g. 300 releases the model after 5min idle (frees VRAM on GPU; next voice message reloads it)
  language: "en"               # GLOBAL language hint for every STT provider (per-provider language wins). Set "" for auto-detect.
  # groq:
  #   model: "whisper-large-v3-turbo"
  #   language: ""             # blank = stt.language > HERMES_LOCAL_STT_LANGUAGE > auto-detect
  openai:
    model: "whisper-1"         # whisper-1 | gpt-4o-mini-transcribe | gpt-4o-transcribe | gpt-transcribe
    language: ""               # auto-detect; set to "en", "es", "fr", etc. to force
  # mistral:
  #   model: "voxtral-mini-latest"  # voxtral-mini-latest | voxtral-mini-2602
  # deepinfra:
  #   # Model id is discovered live from the DeepInfra catalog filtered
  #   # by the `stt` surface tag — leave `model` blank to take the first
  #   # live result. Pin only when you need a specific Whisper variant.
  #   model: ""

# Text-to-speech. Only the deepinfra block is documented here — the
# remaining providers (edge, openai, xai, minimax, mistral, gemini,
# elevenlabs, neutts, kittentts, piper) inherit sensible defaults from
# DEFAULT_CONFIG in hermes_cli/config.py.
# tts:
#   provider: "deepinfra"
#   deepinfra:
#     # Model id is discovered live from the DeepInfra catalog filtered
#     # by the `tts` surface tag — leave `model` blank to take the first
#     # live result.
#     model: ""
#     voice: "default"

# "Hey Hermes" hands-free wake word (off by default; toggle with /wake).
# Full engine/provider options inherit from DEFAULT_CONFIG in
# hermes_cli/config_defaults.py — only capture placement is documented here.
# wake_word:
#   enabled: false
#   capture: "auto"    # auto | local | client
#                      # auto: backend PortAudio mic when one exists; a remote
#                      #       desktop on a mic-less (headless/VPS) backend
#                      #       streams its own mic via the wake.feed RPC.
#                      # local: always the backend mic (wake_word.input_device).
#                      # client: always desktop-streamed PCM (detection stays
#                      #         on the backend).

# Image generation. Each provider plugin reads its own ``image_gen.<name>``
# block; deepinfra discovers models live from
# api.deepinfra.com/v1/openai/models filtered by the ``image-gen`` tag —
# no model id is hardcoded, so retired models disappear automatically.
# image_gen:
#   provider: "deepinfra"
#   deepinfra:
#     # Leave `model` blank for the first live `image-gen`-tagged result.
#     model: ""

# =============================================================================
# Response Pacing (Messaging Platforms)
# =============================================================================
# Add human-like delays between message chunks.
# human_delay:
#   mode: "off"      # "off" | "natural" | "custom"
#   min_ms: 800      # Min delay (custom mode only)
#   max_ms: 2500     # Max delay (custom mode only)

# =============================================================================
# Session Logging
# =============================================================================
# Session trajectories are automatically saved to logs/ directory.
# Each session creates: logs/session_YYYYMMDD_HHMMSS_UUID.json
#
# The session ID is displayed in the welcome banner for easy reference.
# Logs contain full conversation history in trajectory format:
# - System prompt, user messages, assistant responses
# - Tool calls with inputs/outputs
# - Timestamps for debugging
#
# No configuration needed - logging is always enabled.
# To disable, you would need to modify the source code.

# =============================================================================
# Code Execution Sandbox (Programmatic Tool Calling)
# =============================================================================
# The execute_code tool runs Python scripts that call Hermes tools via RPC.
# Intermediate tool results stay out of the LLM's context window.
code_execution:
  timeout: 300         # Max seconds per script before kill (default: 300 = 5 min)
  max_tool_calls: 50   # Max RPC tool calls per execution (default: 50)
  # Local execution uses a persistent session kernel: variables, imports, and
  # loaded data survive across execute_code calls in one conversation, so
  # multi-step data work stops re-loading its inputs every call. Pass
  # reset=true on a call to discard state; a timed-out/interrupted cell kills
  # the kernel (next call starts fresh). Remote terminal backends currently
  # run per-call (remote kernel host is tracked follow-up work). Security
  # scrubbing, tool whitelist, and output redaction are identical either way.
  # kernel_idle_timeout: 1800   # Reap kernels idle longer than this (seconds)
  # max_session_kernels: 4      # Process-wide LRU cap on live kernels

# =============================================================================
# Subagent Delegation
# =============================================================================
# The delegate_task tool spawns child agents with isolated context.
# Supports single tasks and batch mode (default 3 parallel, configurable).
delegation:
  max_iterations: 250                         # Max tool-calling turns per child (default: 250)
  # max_concurrent_children: 10               # Max parallel child agents per batch (default: 10, floor: 1, no ceiling).
                                              # WARNING: values above 10 multiply API cost linearly.
  # max_spawn_depth: 1                        # Delegation tree depth cap (range: 1-3, default: 1 = flat).
                                              # Raise to 2 to allow workers to spawn their own subagents.
                                              # Requires role="orchestrator" on intermediate agents.
  # orchestrator_enabled: true                # Kill switch for role="orchestrator" children (default: true).
  # subagent_auto_approve: false              # When a subagent hits a dangerous-command approval prompt, auto-deny (default: false)
                                              # or auto-approve "once" (true) instead of blocking on stdin.
                                              # The parent TUI owns stdin, so blocking would deadlock; non-interactive resolution is required.
                                              # Both choices emit a logger.warning audit line. Flip to true only for cron/batch pipelines.
  # inherit_mcp_toolsets: true                # When explicit child toolsets are narrowed, also keep the parent's MCP toolsets (default: true). Set false for strict intersection.
  # model: "google/gemini-3-flash-preview"    # Override model for subagents (empty = inherit parent)
  # provider: "openrouter"                    # Override provider for subagents (empty = inherit parent)
  #                                           # Resolves full credentials (base_url, api_key) automatically.
  #                                           # Supported: openrouter, nous, zai, kimi-coding, minimax
  #                                           # Cost tip: keep the parent on a frontier model and pin
  #                                           # delegation.model to an inexpensive one — children carry the
  #                                           # vast majority of tokens, so this is where spend is cut while
  #                                           # planning quality stays with the frontier parent.
  # fallback_providers:                       # Fallback chain for delegated children (same entry format as the
  #   - provider: "openrouter"                # top-level list). Pinned children use only a declared child chain;
  #     model: "deepseek/deepseek-chat"       # unpinned children inherit when unset. [] disables fallback (#65038).

# =============================================================================
# Honcho Integration (Cross-Session User Modeling)
# =============================================================================
# AI-native persistent memory via Honcho (https://honcho.dev/).
# Builds a deeper understanding of the user across sessions and tools.
# Runs alongside USER.md — additive, not a replacement.
#
# Requires: pip install honcho-ai
# Config: ~/.honcho/config.json (shared with Claude Code, Cursor, etc.)
# API key: HONCHO_API_KEY in ~/.hermes/.env or ~/.honcho/config.json
#
# Hermes-specific overrides (optional — most config comes from ~/.honcho/config.json):
# honcho: {}

# =============================================================================
# Display
# =============================================================================
display:
  # Use compact banner mode (hides the ASCII-art banner, shows a single line).
  #   true:  Compact single-line banner
  #   false: Full ASCII banner with tool/skill summary (default)
  compact: false

  # Tool progress display level (CLI and gateway)
  #   off:     Silent — no tool activity shown, just the final response
  #   new:     Show a tool indicator only when the tool changes (skip repeats)
  #   all:     Show every tool call with a short preview (default)
  #   verbose: Full args, results, and debug logs (same as /verbose)
  #   log:     Silent in chat; append every tool call to ~/.hermes/logs/tool_calls.log (gateway only)
  # Toggle at runtime with /verbose in the CLI
  tool_progress: all

  # Per-platform defaults can be quieter than the global setting. Telegram
  # tunes for mobile: tool_progress and busy_ack_detail default off (no
  # per-tool breadcrumb stream, no "iteration 21/60" debug detail in busy
  # acks or heartbeats), but interim_assistant_messages and
  # long_running_notifications STAY ON so the user has real signal between
  # turn start and final answer (mid-turn assistant commentary + a single
  # edit-in-place "⏳ Working — N min" heartbeat). Override under
  # display.platforms.telegram.

  # Auto-cleanup of temporary progress bubbles after the final response lands.
  # On platforms that support message deletion (currently Telegram), this
  # removes the tool-progress bubble, "⏳ Still working..." notices, and
  # context-pressure status messages once the final reply has been delivered —
  # keeping long-running turns visible live, then tidy afterward. Failed runs
  # leave the bubbles in place as breadcrumbs. Off by default.
  # Per-platform override: display.platforms.telegram.cleanup_progress
  #   true:  Delete tracked progress/status bubbles on successful turn
  #   false: Leave everything in place (default)
  # Example:
  #   display:
  #     platforms:
  #       telegram:
  #         cleanup_progress: true
  cleanup_progress: false

  # Natural mid-turn assistant updates.
  # On gateway platforms, when true, completed assistant status messages are
  # sent as separate chat messages. On the Desktop app, when true, mid-turn
  # assistant narration streamed between tool calls is kept in the transcript
  # instead of the bubble collapsing to only the final message on completion.
  # Independent of tool_progress and gateway streaming.
  #   true:  Keep/send mid-turn assistant updates (default)
  #   false: Only keep/send the final response
  interim_assistant_messages: true

  # Gateway-only long-running status heartbeats.
  # When false, the platform does not receive periodic "⏳ Working — N min"
  # notifications even if agent.gateway_notify_interval is non-zero. The
  # heartbeat edits a single message in place (where the adapter supports
  # editing) instead of posting a new bubble each interval.
  # Default: true everywhere, including Telegram (silent agents are worse
  # than a single edit-in-place heartbeat).
  long_running_notifications: true

  # Include detailed iteration/tool/status context in busy acknowledgments
  # and long-running heartbeats. When true, busy acks show "iteration 21/60,
  # terminal, 10 min" and the heartbeat shows "⏳ Working — 12 min,
  # iteration 21/60, terminal". When false (Telegram default), both stay
  # terse: "Interrupting current task" and "⏳ Working — 12 min, terminal".
  busy_ack_detail: true

  # What Enter does when Hermes is already busy (CLI and gateway platforms).
  #   interrupt: Interrupt the current run and redirect Hermes (default)
  #   queue:     Queue your message for the next turn
  #   steer:     Inject your message mid-run via /steer, arriving at the agent
  #              after the next tool call — no interrupt, no role violation.
  #              Falls back to 'queue' if the agent isn't running yet or if
  #              images are attached (steer only carries text).
  # Ctrl+C (or /stop in gateway) always interrupts regardless of this setting.
  # Toggle at runtime with /busy <interrupt|queue|steer>.
  busy_input_mode: interrupt

  # Background process notifications (gateway/messaging only).
  # Controls how chatty the process watcher is when you use
  # terminal(background=true, notify_on_complete=true) from Telegram/Discord/etc.
  #   concise: One-line status message on completion; failures include a short
  #            output tail (default)
  #   off:     No watcher messages at all
  #   result:  Only the final raw-output completion message
  #   error:   Only the final raw-output message when exit code != 0
  #   all:     Running output updates + final raw-output message
  background_process_notifications: concise


  # Play terminal bell when agent finishes a response.
  # Useful for long-running tasks — your terminal will ding when the agent is done.
  # Works over SSH. Most terminals can be configured to flash the taskbar or play a sound.
  #   true:  Ring the terminal bell on each response
  #   false: Silent (default)
  bell_on_complete: false

  # Play terminal bell when a blocking prompt opens and waits on you:
  # clarify questions, dangerous-command approvals, sudo password, secret
  # capture. Same mechanism as bell_on_complete (\a) — works over SSH.
  #   true:  Ring whenever the agent is waiting for your input
  #   false: Silent (default)
  bell_on_prompt: false

  # Show model reasoning/thinking before each response.
  # When enabled, a dim box shows the model's thought process above the response.
  # Toggle at runtime with /reasoning show or /reasoning hide.
  #   true:  Show the reasoning box
  #   false: Hide reasoning (default)
  show_reasoning: false

  # Stream tokens to the terminal as they arrive instead of waiting for the
  # full response. The response box opens on first token and text appears
  # line-by-line. Tool calls are still captured silently.
  #   true:  Stream tokens as they arrive (default)
  #   false: Wait for the full response before rendering
  streaming: true

  # Show [HH:MM] timestamps on user input and assistant response labels.
  # timestamps: false

  # ───────────────────────────────────────────────────────────────────────────
  # Skin / Theme
  # ───────────────────────────────────────────────────────────────────────────
  # Customize CLI visual appearance — banner colors, spinner faces, tool prefix,
  # response box label, and branding text. Change at runtime with /skin <name>.
  #
  # Built-in skins:
  #   default        — Classic Hermes gold/kawaii
  #   ares           — Crimson/bronze war-god theme with spinner wings
  #   mono           — Clean grayscale monochrome
  #   slate          — Cool blue developer-focused
  #   daylight       — Bright light-mode theme
  #   warm-lightmode — Warm paper-tone light-mode theme
  #   poseidon       — Sea-green/teal Olympian theme
  #   sisyphus       — Earthy stone-and-moss theme
  #   charizard      — Fiery orange dragon theme
  #
  # Custom skins: drop a YAML file in ~/.hermes/skins/<name>.yaml
  # Schema (all fields optional, missing values inherit from default):
  #
  #   name: my-theme
  #   description: Short description
  #   colors:
  #     banner_border: "#HEX"    # Panel border
  #     banner_title: "#HEX"     # Panel title
  #     banner_accent: "#HEX"    # Section headers (Available Tools, etc.)
  #     banner_dim: "#HEX"       # Dim/muted text
  #     banner_text: "#HEX"      # Body text (tool names, skill names)
  #     ui_accent: "#HEX"        # UI accent color
  #     response_border: "#HEX"  # Response box border color
  #   spinner:
  #     waiting_faces: ["(⚔)", "(⛨)"]       # Faces shown while waiting
  #     thinking_faces: ["(⚔)", "(⌁)"]      # Faces shown while thinking
  #     thinking_verbs: ["forging", "plotting"]  # Verbs for spinner messages
  #     wings:                                # Optional left/right spinner decorations
  #       - ["⟪⚔", "⚔⟫"]
  #       - ["⟪▲", "▲⟫"]
  #   branding:
  #     agent_name: "My Agent"               # Banner title and branding
  #     welcome: "Welcome message"           # Shown at CLI startup
  #     response_label: " ⚔ Agent "         # Response box header label
  #     prompt_symbol: "⚔"                  # Prompt symbol (bare token; renderers add trailing space)
  #   tool_prefix: "╎"                       # Tool output line prefix (default: ┊)
  #
  skin: default

# =============================================================================
# Model Aliases — short names for /model command
# =============================================================================
# Map short aliases to exact (model, provider, base_url) tuples.
# Used by /model tab completion and resolve_alias().
# Aliases are checked BEFORE the models.dev catalog, so they can route
# to endpoints not in the catalog (e.g. Ollama Cloud, local servers).
#
# An alias pointing at its own endpoint can carry that endpoint's
# credential with `api_key` (a literal, or a "${VAR}" reference) or
# `key_env` (an env var name). `api_key` wins if both are set. When
# neither is set the key is resolved from the alias HOST — never from
# whatever provider was active before the switch, which would send that
# provider's secret to an unrelated third party.
#
# model_aliases:
#   opus:
#     model: claude-opus-4-6
#     provider: anthropic
#   qwen:
#     model: "qwen3.5:397b"
#     provider: custom
#     base_url: "https://ollama.com/v1"
#   glm:
#     model: glm-4.7
#     provider: custom
#     base_url: "https://ollama.com/v1"
#   theta:
#     model: theta-1
#     provider: custom
#     base_url: "https://theta.example.com/v1"
#     key_env: THETA_API_KEY               # or: api_key: "${THETA_API_KEY}"

# =============================================================================
# Privacy
# =============================================================================
# privacy:
#   # Redact PII from the LLM context prompt.
#   # When true, phone numbers are stripped and user/chat IDs are replaced
#   # with deterministic hashes before being sent to the model.
#   # Names and usernames are NOT affected (user-chosen, publicly visible).
#   # Routing/delivery still uses the original values internally.
#   redact_pii: false

# =============================================================================
# Shell-script hooks
# =============================================================================
# Register shell scripts as plugin-hook callbacks.  Each entry is executed as
# a subprocess (shell=False, shlex.split) with a JSON payload on stdin.  On
# stdout the script may return JSON that either blocks the tool call or
# injects context into the next LLM call.
#
# Valid events (mirror hermes_cli.plugins.VALID_HOOKS):
#   pre_tool_call, post_tool_call, pre_llm_call, post_llm_call,
#   pre_api_request, post_api_request, on_session_start, on_session_end,
#   on_session_finalize, on_session_reset, subagent_stop
#
# First-use consent: each (event, command) pair prompts once on a TTY, then
# is persisted to ~/.hermes/shell-hooks-allowlist.json.  Non-interactive
# runs (gateway, cron) need --accept-hooks, HERMES_ACCEPT_HOOKS=1, or the
# hooks_auto_accept key below.
#
# See website/docs/user-guide/features/hooks.md for the full JSON wire
# protocol and worked examples.
#
# hooks:
#   pre_tool_call:
#     - matcher: "terminal"
#       command: "~/.hermes/agent-hooks/block-rm-rf.sh"
#       timeout: 10
#   post_tool_call:
#     - matcher: "write_file|patch"
#       command: "~/.hermes/agent-hooks/auto-format.sh"
#   pre_llm_call:
#     - command: "~/.hermes/agent-hooks/inject-cwd-context.sh"
#   subagent_stop:
#     - command: "~/.hermes/agent-hooks/log-orchestration.sh"
#
# hooks_auto_accept: false


# =============================================================================
# Telemetry
# =============================================================================
# Shared metrics are disabled by default. When enabled, Hermes writes only
# allowlisted aggregate counters and immutable JSON
# packages under $HERMES_HOME/telemetry/shared_metrics.
# Packages include a random profile-scoped ID that stays stable until this
# directory is deleted. It is not derived from hardware, account, or host data.
# Successfully exported local history is retained for 30 days; pending deltas
# are retained until they can be exported.
# This profile-owned choice is not overridden by managed-scope configuration.
#
# Nothing is uploaded unless you also set `send: true`. That is a separate
# opt-in and requires `enabled`; it never turns collection on by itself.
# When sending is on:
#   * only packages whose entire collection period falls within one
#     continuous recorded consent window are ever sent. Consent windows open
#     when you enable sending and close when you disable it, so data
#     collected before you opted in — or during any gap between opt-ins —
#     stays on this machine;
#   * each package carries the profile-scoped ID as-is. It is a random UUID
#     with no hardware, account, or host-derived content, and deleting the
#     shared-metrics directory resets it.
# See docs/observability/relay-shared-metrics.md (Appendix A) for the full
# consent, identity, retention, and deletion decisions.
telemetry:
  shared_metrics:
    enabled: false
    send: false
    # endpoint: https://telemetry.nousresearch.com/v1/telemetry


# =============================================================================
# Update Behavior
# =============================================================================
updates:
  # Disable passive CLI/banner update checks with `hermes config set updates.check false`.
  # Explicit `hermes update --check` and `hermes update` remain available.
  check: true
  # Create a full HERMES_HOME zip before every `hermes update`.
  # Backups land in ~/.hermes/backups/ and can be restored with `hermes import`.
  # Off by default because large homes can add minutes to every update.
  pre_update_backup: false

  # Number of pre-update backup zips to retain.
  backup_keep: 5

  # What non-interactive updates do with local source edits in the Hermes repo.
  # Interactive terminal updates always prompt before restoring the autostash.
  #
  #   stash   - auto-stash before pull, then auto-restore after success (default)
  #   discard - drop the update-created stash after success; use only on managed
  #             installs where local source edits should not persist
  non_interactive_local_changes: "stash"


# =============================================================================
# Web Dashboard
# =============================================================================
# OAuth gate configuration for `hermes dashboard --host <non-loopback>`.
# The bundled Nous Portal plugin reads these on startup; settings here are
# the canonical surface. Each can be overridden by an environment variable:
#
#   dashboard.oauth.client_id   <-  HERMES_DASHBOARD_OAUTH_CLIENT_ID
#   dashboard.oauth.portal_url  <-  HERMES_DASHBOARD_PORTAL_URL
#   dashboard.public_url        <-  HERMES_DASHBOARD_PUBLIC_URL
#
# Env wins when set to a non-empty value. This is what Fly.io's platform-
# secret injection uses to push per-deploy client_ids without needing to
# bake a config.yaml into the image. Empty env values are treated as unset
# so a provisioned-but-not-populated secret can't shadow a valid entry here.
#
# Local dev / on-prem deploys should typically set these via config.yaml
# (the ~/.hermes/.env file is reserved for API keys and secrets).
#
# dashboard:
#   oauth:
#     client_id: ""    # agent:{instance_id}; Portal provisions this at deploy
#     portal_url: ""   # blank → default https://portal.nousresearch.com
#
#   # Force the absolute base URL the OAuth callback (and any other public
#   # URL the dashboard hands to external systems) is built from. Set this
#   # for deploys behind reverse proxies that don't reliably forward
#   # X-Forwarded-Host / X-Forwarded-Proto / X-Forwarded-Prefix (manual
#   # nginx setups, on-prem ingresses, custom-domain Fly deploys without
#   # full proxy header chains).
#   #
#   # When set, the value is the complete authority: scheme + host +
#   # optional path prefix (e.g. "https://example.com/hermes"). The OAuth
#   # callback URL becomes "<public_url>/auth/callback" — X-Forwarded-Prefix
#   # is IGNORED on this code path because the operator has explicitly
#   # declared the public URL and we no longer need to guess.
#   #
#   # Leave empty to use the existing proxy-header reconstruction (the
#   # default — works on Fly.io out of the box).
#   #
#   #   public_url: "https://example.com/hermes"
#   #
#   # Reverse proxies connecting from another container or host are not
#   # trusted by default. Add only the proxy's exact IP address, or a bounded
#   # CIDR for a dedicated proxy network, so X-Forwarded-Proto and
#   # X-Forwarded-For can be honored. Loopback is always trusted. Wildcards
#   # and /0 networks are rejected.
#   #
#   #   trusted_proxies:
#   #     - "172.20.0.5"
#   #     # - "172.20.0.0/24"  # dedicated network, if the IP is dynamic
#
# -----------------------------------------------------------------------------
# Self-hosted OIDC dashboard auth (generic OpenID Connect — Authentik,
# Keycloak, Zitadel, Authelia, Auth0, Okta, Google, …). Use this INSTEAD of the
# nous block above when gating the dashboard with your own identity provider.
# Each setting can be overridden by an environment variable:
#
#   dashboard.oauth.self_hosted.issuer         <-  HERMES_DASHBOARD_OIDC_ISSUER
#   dashboard.oauth.self_hosted.client_id      <-  HERMES_DASHBOARD_OIDC_CLIENT_ID
#   dashboard.oauth.self_hosted.scopes         <-  HERMES_DASHBOARD_OIDC_SCOPES
#   dashboard.oauth.self_hosted.client_secret  <-  HERMES_DASHBOARD_OIDC_CLIENT_SECRET
#
# dashboard:
#   oauth:
#     provider: self-hosted
#     self_hosted:
#       issuer: "https://auth.example.com/application/o/hermes/"  # required
#       client_id: "hermes-dashboard"                             # required
#       scopes: "openid profile email"                            # optional
#
#       # OPTIONAL — set ONLY if your IDP registered the client as
#       # *confidential* (Authentik / Keycloak often default to this). When
#       # set, Hermes authenticates the client at the token endpoint
#       # (client_secret_basic or client_secret_post, auto-selected from the
#       # IDP's discovery doc) IN ADDITION to PKCE. Leave unset for a public
#       # (PKCE-only) client — the common case.
#       #
#       # This is a CREDENTIAL: prefer setting HERMES_DASHBOARD_OIDC_CLIENT_SECRET
#       # in ~/.hermes/.env over putting it here in config.yaml.
#       #   client_secret: ""

# =============================================================================
# External secret sources
# =============================================================================
# Pull provider credentials from external secret managers at process startup
# instead of storing them in ~/.hermes/.env. Only the manager's bootstrap
# credential (e.g. BWS_ACCESS_TOKEN / OP_SERVICE_ACCOUNT_TOKEN) lives in .env
# (or your shell / desktop session); everything else rotates centrally.
# Failures never block startup — Hermes warns once and continues with
# whatever .env already had.
#
# Multiple sources can be enabled at once:
#   - "mapped" sources (explicit VAR -> ref bindings, e.g. 1Password's env:
#     map) beat "bulk" sources (whole-project dumps like Bitwarden BSM)
#   - within a shape, the first source to claim a var wins; later claims
#     are skipped with a startup warning (never a silent clobber)
#   - a source's override_existing lets it beat .env/shell values, but
#     never another secret source's claim
# Docs: https://hermes-agent.nousresearch.com/docs/user-guide/secrets/
#
# secrets:
#   # Optional explicit ordering of enabled sources.
#   # sources: [onepassword, bitwarden]
#
#   # ---- Bitwarden Secrets Manager (bws CLI) --------------------------------
#   bitwarden:
#     enabled: false
#     access_token_env: BWS_ACCESS_TOKEN   # bootstrap token, sourced from .env
#     project_id: ""                       # UUID of the BSM project to sync
#     server_url: ""                       # "" = US Cloud; EU/self-hosted URL otherwise
#     cache_ttl_seconds: 300               # 0 disables fresh caching
#     encrypted_cache:                     # optional encrypted stale fallback
#       enabled: false
#       max_stale_seconds: 0               # 0 disables stale fallback
#     override_existing: true              # BSM values win over existing env
#     auto_install: true                   # lazy-download bws into ~/.hermes/bin
#
#   # ---- 1Password (op CLI) -------------------------------------------------
#   onepassword:
#     enabled: false
#     # Map env-var names to op:// secret references. Each is resolved with a
#     # single `op read` at startup.
#     env:
#       OPENAI_API_KEY: "op://Private/OpenAI/api key"
#       ANTHROPIC_API_KEY: "op://Private/Anthropic/credential"
#     account: ""                                   # op --account shorthand; "" = default
#     service_account_token_env: OP_SERVICE_ACCOUNT_TOKEN  # headless auth; unset = desktop session
#     binary_path: ""                               # "" = resolve op via PATH; else absolute path
#     cache_ttl_seconds: 300                        # 0 disables BOTH cache layers
#     override_existing: true                       # resolved values win over existing env
#
#   # ---- Command helper (any CLI vault) --------------------------------------
#   # Run a user-configured helper that prints KEY=VALUE lines on stdout —
#   # works with any secret store that has a CLI: keepassxc-cli, secret-tool,
#   # pass, gpg, or a script that cats a tmpfs env file.  Composes with the
#   # sources above (enable any combination).  POSIX-only (needs /bin/sh).
#   # The helper must be fast and NON-interactive (hard timeout, 1 MiB cap);
#   # its stderr is discarded so diagnostics can't leak secret material.
#   command:
#     enabled: false
#     command: "cat /run/user/1000/hermes-secrets.env"
#     helper_timeout_seconds: 3
#     override_existing: false             # .env/shell win by default
#   # This runs ONCE per process at startup, so it cannot replace a credential
#   # that expires mid-session. For a single provider whose token needs
#   # re-minting during a session, use `providers.<name>.key_cmd` instead.