PR #17405: feat(honcho): add context injection section controls

• Repository: NousResearch/hermes-agent
• Author: de1tydev
• State: open | merged: False
• Link: https://github.com/NousResearch/hermes-agent/pull/17405

Description (problem / solution / changelog)

What does this PR do?

Adds contextInjection controls for the Honcho memory provider so users can choose which base-context sections are formatted into the auto-injected prompt.

This is useful for hybrid / context recall modes where users want automatic Honcho context, but do not want every session/user/AI representation block injected into the model context on every refresh.

The new config is backward compatible: all sections default to true, preserving existing behavior.

Example:

{
  "contextInjection": {
    "sessionSummary": true,
    "userRepresentation": false,
    "userPeerCard": true,
    "aiRepresentation": false,
    "aiPeerCard": false
  }
}

contextInjection can be configured at the root or under hosts.<host>. Host values override root values per key.

This only controls formatted base-context injection. It does not disable Honcho tools, message saving, dialectic supplements, or necessarily underlying prefetch calls. Users who want no automatic injection should continue to use recallMode: "tools".

Related Issue

Resolves #17412

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

• plugins/memory/honcho/client.py
  • Adds contextInjection defaults and parsing.
  • Supports root + host-level config with host per-key overrides.
  • Ignores unknown keys and non-object values safely.
• plugins/memory/honcho/__init__.py
  • Applies section-level gates when formatting first-turn/base Honcho context.
• tests/test_honcho_client_config.py
  • Adds config parsing coverage.
• tests/honcho_plugin/test_session.py
  • Adds formatter behavior coverage.
• Honcho docs updated with config examples and semantics.

How to Test

python -m py_compile plugins/memory/honcho/client.py plugins/memory/honcho/__init__.py
python -m pytest tests/honcho_plugin tests/test_honcho_client_config.py -q -o 'addopts='
git diff --check origin/main..HEAD

Result locally:

283 passed, 1 warning

Checklist

Code

• I've read the Contributing Guide
• My commit messages follow Conventional Commits
• I searched for existing PRs to make sure this isn't a duplicate
• My PR contains only changes related to this feature
• I've run pytest tests/ -q and all tests pass
• I've added tests for my changes
• I've tested on my platform: Ubuntu/Linux, Python 3.12

Documentation & Housekeeping

• I've updated relevant documentation
• I've updated cli-config.yaml.example if I added/changed config keys — N/A
• I've updated CONTRIBUTING.md or AGENTS.md if I changed architecture or workflows — N/A
• I've considered cross-platform impact — config parsing / prompt formatting only
• I've updated tool descriptions/schemas if I changed tool behavior — N/A

Changed files

• optional-skills/autonomous-ai-agents/honcho/SKILL.md (modified, +34/-3)
• plugins/memory/honcho/README.md (modified, +33/-1)
• plugins/memory/honcho/__init__.py (modified, +9/-5)
• plugins/memory/honcho/client.py (modified, +31/-0)
• tests/honcho_plugin/test_session.py (modified, +89/-0)
• tests/test_honcho_client_config.py (modified, +111/-0)
• website/docs/user-guide/features/memory-providers.md (modified, +15/-0)