fix(skills): avoid mislabeling local skills as builtin in skills list by arceus77-7 · Pull Request #862 · NousResearch/hermes-agent

arceus77-7 · 2026-03-10T21:23:33Z

What does this PR do?

Fixes incorrect source labeling in hermes skills list.

Previously, every non-hub-installed skill was shown as builtin. This misclassified user-managed skills (for example skills synced in from external managers) as built-in Hermes skills.

This PR classifies installed skills into:

hub (present in Hub lockfile)
builtin (present in bundled skills manifest)
local (everything else under ~/.hermes/skills/)

It also adds a --source local filter.

Related Issue

Fixes #861

Type of Change

🐛 Bug fix (non-breaking change that fixes an issue)
✅ Tests (adding or improving test coverage)

Changes Made

Updated source classification in hermes_cli/skills_hub.py (do_list):
- Reads bundled skill names from tools.skills_sync._read_manifest().
- Distinguishes hub, builtin, and local skills.
- Updates summary counts to include local skills.
Updated /skills help text to document --source hub|builtin|local.
Updated CLI argument choices in hermes_cli/main.py to include local.
Added tests in tests/hermes_cli/test_skills_hub.py:
- test_do_list_distinguishes_hub_builtin_and_local
- test_do_list_local_filter

How to Test

Run tests:
- ./venv/bin/pytest tests/hermes_cli/test_skills_hub.py -q
Add or ensure a non-bundled local skill exists under ~/.hermes/skills/.
Run:
- hermes skills list
- hermes skills list --source local
Verify local skill shows Source=local and no longer appears as builtin.

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 added tests for my changes (required for bug fixes, strongly encouraged for features)
I've tested on my platform: macOS

Documentation & Housekeeping

N/A for this scoped bugfix: README/docs/config/AGENTS/tool-schema updates
I've considered cross-platform impact (Windows, macOS) per the compatibility guide

Screenshots / Logs

./venv/bin/pytest tests/hermes_cli/test_skills_hub.py -q -> 3 passed
Note: full pytest tests/ -q was not run for this focused fix.

Follow-up to PR #862 (local skills classification by arceus77-7): - Remove unnecessary isinstance guard on _read_manifest() return value — it always returns Dict[str, str], so set() on it suffices. - Extract repeated hub-dir monkeypatching into a shared pytest fixture (hub_env). - Add three_source_env fixture for source-classification tests. - Add _read_manifest monkeypatch to test_do_list_initializes_hub_dir (was fragile — relied on empty skills list masking the real manifest). - Add test coverage for --source hub and --source builtin filters. - Extract _capture() helper to reduce console/StringIO boilerplate. 5 tests, all green.

chore(skills): clean up PR #862 + feat(docs): add search to Docusaurus

…ource-provenance Merging — clean fix for local skills mislabeling. Follow-up cleanup coming.

…, DRY up tests Follow-up to PR NousResearch#862 (local skills classification by arceus77-7): - Remove unnecessary isinstance guard on _read_manifest() return value — it always returns Dict[str, str], so set() on it suffices. - Extract repeated hub-dir monkeypatching into a shared pytest fixture (hub_env). - Add three_source_env fixture for source-classification tests. - Add _read_manifest monkeypatch to test_do_list_initializes_hub_dir (was fragile — relied on empty skills list masking the real manifest). - Add test coverage for --source hub and --source builtin filters. - Extract _capture() helper to reduce console/StringIO boilerplate. 5 tests, all green.

…c877bdeb chore(skills): clean up PR NousResearch#862 + feat(docs): add search to Docusaurus