Add global config switch to disable automatic session naming by baxen · Pull Request #7052 · block/goose

baxen · 2026-02-06T18:28:17Z

Adds a GOOSE_DISABLE_SESSION_NAMING configuration flag that lets users turn off the LLM-powered automatic session naming feature. This is useful in headless/CI environments or for users who find the background naming calls unnecessary, reducing token usage and latency.

Copilot

Pull request overview

Adds a global configuration switch to disable the LLM-powered automatic session naming, intended to reduce unnecessary background LLM calls (tokens/latency) in CI/headless usage.

Changes:

Introduces GOOSE_DISABLE_SESSION_NAMING as a new config/env value.
Short-circuits SessionManager::maybe_update_name when the flag is enabled.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 5 comments.

File	Description
crates/goose/src/session/session_manager.rs	Adds an early-return gate in `maybe_update_name` based on the new global config flag.
crates/goose/src/config/base.rs	Registers `GOOSE_DISABLE_SESSION_NAMING` via `config_value!` to generate get/set accessors.

Copilot · 2026-02-06T18:32:49Z

crates/goose/src/session/session_manager.rs

+        if Config::global()
+            .get_goose_disable_session_naming()
+            .unwrap_or(false)
+        {
+            return Ok(());
+        }


Calling Config::global().get_goose_disable_session_naming() here will read (and potentially create/migrate) the config file on every maybe_update_name invocation, which is likely on every agent reply; consider checking an env var first and/or caching this flag (or moving the check to the caller) to avoid repeated disk I/O and latency.

Copilot · 2026-02-06T18:32:49Z

crates/goose/src/session/session_manager.rs

+        if Config::global()
+            .get_goose_disable_session_naming()
+            .unwrap_or(false)
+        {


Because this flag is defined as a bool, common CI-style values like GOOSE_DISABLE_SESSION_NAMING=1 will fail to deserialize and then be silently treated as false due to unwrap_or(false); consider treating this as a presence-based flag (like GOOSE_DISABLE_KEYRING) or otherwise accepting "1"/"0" to avoid a confusing no-op configuration.

Copilot · 2026-02-06T18:32:50Z

crates/goose/src/session/session_manager.rs

    pub async fn maybe_update_name(&self, id: &str, provider: Arc<dyn Provider>) -> Result<()> {
+        if Config::global()
+            .get_goose_disable_session_naming()
+            .unwrap_or(false)
+        {
+            return Ok(());
+        }


The new disable-switch behavior in maybe_update_name isn’t covered by tests in this file; adding a unit/integration test that sets the flag and asserts generate_session_name is not invoked would prevent regressions (especially since this runs in a background task).

Copilot · 2026-02-06T18:32:50Z

crates/goose/src/config/base.rs

 config_value!(GOOSE_MODEL, String);
 config_value!(GOOSE_PROMPT_EDITOR, Option<String>);
 config_value!(GOOSE_MAX_ACTIVE_AGENTS, usize);
+config_value!(GOOSE_DISABLE_SESSION_NAMING, bool);


This new config/env key should be added to the user-facing configuration docs (both the environment variable list and the config.yaml settings list), otherwise users won’t discover it or may assume unsupported values (e.g., "1") work.

Copilot · 2026-02-06T18:32:50Z

crates/goose/src/config/base.rs

 config_value!(GOOSE_MODEL, String);
 config_value!(GOOSE_PROMPT_EDITOR, Option<String>);
 config_value!(GOOSE_MAX_ACTIVE_AGENTS, usize);
+config_value!(GOOSE_DISABLE_SESSION_NAMING, bool);


Declaring GOOSE_DISABLE_SESSION_NAMING as bool means env-var values like "1"/"yes" won’t deserialize to bool via get_param, which is inconsistent with other presence-based disable flags (e.g., GOOSE_DISABLE_KEYRING); consider using a presence-based string flag or broadening accepted boolean-like values.

Suggested change

config_value!(GOOSE_DISABLE_SESSION_NAMING, bool);

config_value!(GOOSE_DISABLE_SESSION_NAMING, String);

DOsinga

Shouldn't we be doing this at the calling side of things in agent:reply_internal? there we spawn to do this on the side, maybe best to not do that at all?

Copilot

Pull request overview

Copilot reviewed 4 out of 4 changed files in this pull request and generated 4 comments.

Copilot · 2026-02-06T22:38:53Z

crates/goose/src/providers/codex.rs

        let skip_git_check = config
            .get_codex_skip_git_check()
-            .map(|s| s.to_lowercase() == "true")
+            .map(|s| is_truthy(&s))
            .unwrap_or(false);


Because get_codex_skip_git_check() is backed by a String-valued config key, common env values like CODEX_SKIP_GIT_CHECK=true/false/1/0 will be parsed as JSON primitives and fail to deserialize into String, causing this to fall back to false and effectively ignore the setting; consider making the config key typed as bool (and read it directly) so env/config-file booleans work as expected.

Copilot · 2026-02-06T22:38:53Z

crates/goose/src/config/base.rs

 config_value!(GOOSE_MODEL, String);
 config_value!(GOOSE_PROMPT_EDITOR, Option<String>);
 config_value!(GOOSE_MAX_ACTIVE_AGENTS, usize);
+config_value!(GOOSE_DISABLE_SESSION_NAMING, String, "false");


GOOSE_DISABLE_SESSION_NAMING is defined as a String with a default, but Config::get_param parses env/config-file true/false as booleans, so setting GOOSE_DISABLE_SESSION_NAMING=true (or false) will fail to deserialize into String and the flag won’t work; define this config value as a bool (and default via unwrap_or(false) on NotFound) or adjust the config parsing so string-typed keys can still accept boolean literals.

Suggested change

config_value!(GOOSE_DISABLE_SESSION_NAMING, String, "false");

config_value!(GOOSE_DISABLE_SESSION_NAMING, bool, false);

Copilot · 2026-02-06T22:38:54Z

crates/goose/src/config/base.rs

+/// Interpret a string value as a boolean in a permissive, presence-based way.
+/// Returns true for values like "true", "1", "yes", "on" (case-insensitive).
+/// Returns false for "false", "0", "no", "off", or empty strings.
+pub fn is_truthy(value: &str) -> bool {
+    !matches!(
+        value.trim().to_lowercase().as_str(),
+        "false" | "0" | "no" | "off" | ""
+    )
+}


is_truthy claims to accept values like "true" and "1", but for env/config entries those values are currently parsed into boolean/number types before deserialization, so string-backed config keys will never see them and the behavior will be surprising; either scope this helper to raw strings that are known to stay strings (e.g., "yes"/"on"), or switch the relevant config keys to typed bool so "true"/"1" work as documented.

Copilot · 2026-02-06T22:38:54Z

crates/goose/src/agents/agent.rs

+            .get_goose_disable_session_naming()
+            .map(|v| is_truthy(&v))


This uses a String-backed config value plus is_truthy, but with the current env/config parsing GOOSE_DISABLE_SESSION_NAMING=true/false will deserialize as a bool and error, so unwrap_or(false) will ignore the user’s intent; prefer a bool-typed config getter here (e.g., get_param::<bool> or a config_value! bool) so the switch behaves predictably.

Suggested change

.get_goose_disable_session_naming()

.map(|v| is_truthy(&v))

.get_param::<bool>("goose_disable_session_naming")

codefromthecrypt

since we can't predict the order of the session request, this is the single complicating thing of VCR style replay tests. Also this complicates the interim solution. I'd like to thread a config on ACP side to not require the Config::global() singleton to accomplish this.

Copilot

Pull request overview

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

Copilot · 2026-02-06T23:29:03Z

crates/goose/src/agents/agent.rs

+        if !naming_disabled {
+            let manager_for_spawn = session_manager.clone();
+            tokio::spawn(async move {
+                if let Err(e) = manager_for_spawn
+                    .maybe_update_name(&session_id, provider)
+                    .await
+                {
+                    warn!("Failed to generate session description: {}", e);
+                }
+            });
+        }


This new config-gated branch changes whether a background naming task is spawned, but there’s no test coverage validating that session naming is skipped when the flag is enabled; add a unit/integration test that sets GOOSE_DISABLE_SESSION_NAMING and asserts maybe_update_name is not invoked.

codefromthecrypt · 2026-02-07T00:39:43Z

#7062 follow-up so it is used in acp and scenario tests

* 'main' of github.com:block/goose: refactor: move disable_session_naming into AgentConfig (#7062) Add global config switch to disable automatic session naming (#7052) docs: add blog post - 8 Things You Didn't Know About Code Mode (#7059) fix: ensure animated elements are visible when prefers-reduced-motion is enabled (#7047) Show recommended model on failture (#7040) feat(ui): add session content search via API (#7050) docs: fix img url (#7053) Desktop UI for deleting custom providers (#7042) Add blog post: How I Used RPI to Build an OpenClaw Alternative (#7051)

)

Copilot AI review requested due to automatic review settings February 6, 2026 18:28

Copilot started reviewing on behalf of baxen February 6, 2026 18:28 View session

Copilot AI reviewed Feb 6, 2026

View reviewed changes

alexhancock approved these changes Feb 6, 2026

View reviewed changes

DOsinga reviewed Feb 6, 2026

View reviewed changes

Copilot AI review requested due to automatic review settings February 6, 2026 22:35

Copilot started reviewing on behalf of baxen February 6, 2026 22:35 View session

Copilot AI reviewed Feb 6, 2026

View reviewed changes

codefromthecrypt approved these changes Feb 6, 2026

View reviewed changes

codefromthecrypt mentioned this pull request Feb 6, 2026

fix: propagate session ID to LLM request headers in rename task #5624

Closed

10 tasks

Copilot AI review requested due to automatic review settings February 6, 2026 23:25

baxen force-pushed the baxen/no-names branch from 0857aa0 to aabb583 Compare February 6, 2026 23:25

Copilot started reviewing on behalf of baxen February 6, 2026 23:26 View session

feat(config): add global switch to disable session naming

37e3cab

baxen force-pushed the baxen/no-names branch from aabb583 to 37e3cab Compare February 6, 2026 23:26

baxen enabled auto-merge February 6, 2026 23:28

Copilot AI reviewed Feb 6, 2026

View reviewed changes

baxen added this pull request to the merge queue Feb 6, 2026

Merged via the queue into main with commit 96f903d Feb 6, 2026
16 checks passed

baxen deleted the baxen/no-names branch February 6, 2026 23:35

codefromthecrypt mentioned this pull request Feb 7, 2026

refactor: move disable_session_naming into AgentConfig #7062

Merged

3 tasks

kuccello pushed a commit to kuccello/goose that referenced this pull request Feb 7, 2026

Add global config switch to disable automatic session naming (block#7052

5163ca5

)

	config_value!(GOOSE_DISABLE_SESSION_NAMING, bool);
	config_value!(GOOSE_DISABLE_SESSION_NAMING, String);

	.get_goose_disable_session_naming()
	.map(\|v\| is_truthy(&v))
	.get_param::<bool>("goose_disable_session_naming")

Conversation

baxen commented Feb 6, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Feb 6, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 6, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 6, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 6, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 6, 2026

Choose a reason for hiding this comment

Uh oh!

DOsinga left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Feb 6, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 6, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 6, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 6, 2026

Choose a reason for hiding this comment

Uh oh!

codefromthecrypt left a comment

Choose a reason for hiding this comment

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Feb 6, 2026

Choose a reason for hiding this comment

Uh oh!

Uh oh!

codefromthecrypt commented Feb 7, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants