What does this PR do?

Adds a profile-scoped autonomy runtime to the messaging gateway so Hermes can track open-ended watch items, revisit them in the background, and surface useful updates naturally without relying on cron for everything.

This introduces a gateway-only autonomy system with durable state in SQLite, hot-path watch registration for explicit requests, hidden post-turn extraction for implied follow-ups, supervisor-driven findings/drafts, and proactive inbox/outbox delivery to a configured home chat. It also adds config controls for extraction mode, proactivity level, quiet hours, retention, and tool access, plus docs for setup and behaviour.

Type of Change

• 🐛 Bug fix (non-breaking change that fixes an issue)
• ✨ New feature (non-breaking change that adds functionality)
• 🔒 Security fix
• 📝 Documentation update
• ✅ Tests (adding or improving test coverage)
• ♻️ Refactor (no behavior change)
• 🎯 New skill (bundled or hub)

Changes Made

• Added profile-scoped gateway autonomy state and runtime in gateway/autonomy.py, gateway/run.py, and hermes_state.py
• Added autonomy config support, including extract_behavior, home chat targeting, quiet hours, retention, and proactivity controls in gateway/config.py and hermes_cli/config.py
• Added new tools for open-ended monitoring and read-only cron inspection in tools/autonomy_watch_tool.py and tools/cron_inspect_tool.py
• Updated tool discovery, tool ordering, and gateway prompt guidance so open-ended monitoring prefers autonomy over cron in model_tools.py, toolsets.py, and run_agent.py
• Added /autonomy command support and autonomy-aware session metadata in hermes_cli/commands.py and gateway/session.py
• Added focused autonomy/config/tool/session tests in tests/gateway/test_autonomy_command.py, tests/gateway/test_autonomy_helpers.py, tests/gateway/test_autonomy_intake_guard.py, tests/tools/test_autonomy_watch_tool.py, tests/tools/test_cron_inspect_tool.py, and related config/state/session tests
• Added implementation/spec handoff doc in docs/profile-autonomy-architecture-spec.md
• Updated user and developer docs, including the new autonomy feature page and config docs, in website/docs/user-guide/features/autonomy.md, website/docs/user-guide/configuration.md, and website/docs/developer-guide/gateway-internals.md

How to Test

1. Enable autonomy in a profile config.yaml, set a home chat, and start the gateway.
2. In a messaging thread, ask Hermes for open-ended monitoring like Keep an eye on OpenAI for me. Let me know if anything interesting happens and verify it creates an autonomy watch instead of inventing a cron schedule.
3. Use /autonomy, /autonomy watch, and /autonomy runs to confirm watch registration, background supervisor activity, and clean run summaries.

Checklist

Code

• I've read the Contributing Guide
• My commit messages follow Conventional Commits (fix(scope):, feat(scope):, etc.)
• I searched for existing PRs to make sure this isn't a duplicate
• My PR contains only changes related to this fix/feature (no unrelated commits)
• I've run pytest tests/ -q and all tests pass (All tests associated with PR are green, failed tests are existing)
• I've added tests for my changes (required for bug fixes, strongly encouraged for features)
• I've tested on my platform: macOS

Documentation & Housekeeping

• I've updated relevant documentation (README, docs/, docstrings) — or N/A
• I've updated cli-config.yaml.example if I added/changed config keys — or N/A
• I've updated CONTRIBUTING.md or AGENTS.md if I changed architecture or workflows — or N/A
• I've considered cross-platform impact (Windows, macOS) per the compatibility guide — or N/A
• I've updated tool descriptions/schemas if I changed tool behavior — or N/A

Screenshots / Logs

Focused verification after merging main:

source venv/bin/activate && python -m pytest \
  tests/gateway/test_autonomy_command.py \
  tests/gateway/test_autonomy_helpers.py \
  tests/gateway/test_autonomy_intake_guard.py \
  tests/gateway/test_config.py \
  tests/gateway/test_session.py \
  tests/test_model_tools.py \
  tests/tools/test_autonomy_watch_tool.py \
  tests/tools/test_cron_inspect_tool.py \
  tests/test_hermes_state.py -q

Result: 251 passed