fix: prefer Claude Code Chrome over agent-browser for browser automation

- Rename agent-browser.md rule to browser-automation.md with Chrome as preferred tool
- Add detection logic: check for mcp__claude-in-chrome__* tools in deferred tools list
- Show fallback warning when Chrome extension is not available
- Update all commands (spec-plan, spec-verify, spec-implement, spec-bugfix-*) with
  browser tool resolution step supporting both Chrome and agent-browser workflows
- Add Chrome Extension section to README, installation docs, and installer post-install
- Add Claude Code Chrome to open-source-tools docs as preferred testing tool
- Update verification.md, testing.md, settings.json references
- Move CodeGraph sync from hooks to launcher wrapper (codegraph.py)
- Refactor installer dependencies step with improved structure
- Clean up development-practices.md exploration tooling section

Author: maxritter

.gitignore

@@ -222,6 +222,7 @@ __marimo__/
 .streamlit/secrets.toml
 
 # Claude Code
+scheduled_tasks.lock
 CLAUDE.md
 AGENTS.md
 spec-annotate

README.md

@@ -73,6 +73,13 @@ Installs globally on macOS, Linux, and Windows (WSL2). All tools and rules go to
 
 </details>
 
+<details>
+<summary><b>Chrome Extension (Recommended)</b></summary>
+
+For the best browser automation and E2E testing experience, install the [Claude Code Chrome extension](https://code.claude.com/docs/en/chrome). Pilot automatically detects it and prefers it over agent-browser. In environments where the extension can't be installed (dev containers, headless CI), Pilot falls back to agent-browser automatically.
+
+</details>
+
 <details>
 <summary><b>Installing a specific version or uninstalling</b></summary>
 

docs/docusaurus/docs/features/rules.md

@@ -27,7 +27,7 @@ Rules load automatically at session start — they're enforced standards, not su
 ### Tools (3 rules)
 
 - `cli-tools.md` — Pilot CLI, Probe code search, RTK token optimization
-- `agent-browser.md` — Browser automation for E2E UI testing
+- `browser-automation.md` — Browser automation for E2E UI testing (Claude Code Chrome preferred, agent-browser fallback)
 - `mcp-servers.md` — MCP server reference and tool selection guidance
 
 ## Coding Standards — Activated by File Type

docs/docusaurus/docs/getting-started/installation.md

@@ -25,11 +25,17 @@ Run from any directory — it installs globally to `~/.pilot/` and `~/.claude/`.
 | 1 | Prerequisites | Checks/installs Homebrew, Node.js, Python 3.12+, uv, git, jq |
 | 2 | Claude files | Sets up `~/.claude/` plugin — rules, commands, hooks, MCP servers |
 | 3 | Config files | Creates `.nvmrc` and project config |
-| 4 | Dependencies | Installs Probe, RTK, CodeGraph, agent-browser, language servers |
+| 4 | Dependencies | Installs Probe, RTK, CodeGraph, agent-browser (fallback), language servers |
 | 5 | Shell integration | Auto-configures bash, fish, and zsh with the `pilot` alias |
 | 6 | VS Code extensions | Installs recommended extensions for your language stack |
 | 7 | Finalize | Success message with next steps |
 
+## Chrome Extension (Recommended)
+
+For the best browser automation and E2E testing experience, install the [Claude Code Chrome extension](https://code.claude.com/docs/en/chrome). It provides richer visual context and more reliable interactions than the agent-browser fallback.
+
+Pilot automatically detects the Chrome extension and uses it when available. In environments where the extension can't be installed (e.g., dev containers, headless CI), Pilot falls back to agent-browser automatically.
+
 ## Dev Container
 
 Pilot Shell works inside Dev Containers. Copy the `.devcontainer` folder from the [Pilot Shell repository](https://github.com/maxritter/pilot-shell/tree/main/.devcontainer) into your project, adapt it to your needs (base image, extensions, dependencies), and run the installer inside the container. The installer auto-detects the container environment and skips system-level dependencies like Homebrew.

docs/docusaurus/docs/workflows/spec.md

@@ -67,7 +67,7 @@ Investigation-first flow for targeted fixes. Finds the root cause before touchin
 - Full test suite + type checking + lint + build verification
 - Features: unified review sub-agent (optional, enabled by default)
 - Bugfixes: regression test + full suite — no sub-agents needed
-- For UI features: executes the plan's **E2E test scenarios** step-by-step via browser automation — tracks pass/fail per scenario, auto-fixes failures (up to 2 attempts), escalates persistent failures to known issues; results written back to the plan file
+- For UI features: executes the plan's **E2E test scenarios** step-by-step via browser automation (Claude Code Chrome preferred, agent-browser as fallback) — tracks pass/fail per scenario, auto-fixes failures (up to 2 attempts), escalates persistent failures to known issues; results written back to the plan file
 - Auto-fixes findings, loops back until all checks pass
 - After automated checks pass, prompts you to **review code changes** in the Console's Changes tab — enable Review mode, add inline annotations on any diff line (they save automatically), and the agent addresses them before marking the spec as verified