Repository navigation
Development
| Path | Holds |
|---|---|
.claude-plugin/plugin.json, marketplace.json
|
the plugin manifest and its one-plugin marketplace |
skills/shieldwright/SKILL.md |
the skill |
skills/shieldwright/references/ |
repo-leak, accounts, device, incidents, family |
skills/shieldwright/scripts/sw.py |
the checker, standard library only |
tests/test_sw.py |
unit tests |
evals/ |
skill eval cases; fixtures/make_leak_repo.py builds the leak repository at run time |
ai-docs/ |
handoff, log, plans and research for agents |
python -m pytest
python -m unittest discover -s testsBoth ran 14 tests at 0.1.2. pytest.ini limits collection to tests/, so the eval fixture is not collected. Tests build fake tokens at run time so the repository holds no literal secret, and the gh audit is tested against a fake API.
Skill evals are in evals/evals.json, results in TESTS.md. The skill is an evergreen unit: research in RESEARCH.md, lessons in LEARNINGS.md, state in evergreen.json.
The repository has no CI. Run both test commands, bump version in .claude-plugin/plugin.json, add a CHANGELOG entry, merge, tag vX.Y.Z on main and create the GitHub release. Then update this wiki's Versions and upgrading page and footer.
See SECURITY.md.
This wiki describes shieldwright 0.1.2 and was last updated on 2026-11-21. The plugin is MIT licensed. Report problems in the issues. Source and releases: m4bwav/shieldwright.