Tree
Skill CongBao/skill-steward/packages/integrations/tests/fixtures/companion-legacy/alpha-0.3.0-alpha.1/tree
Cross-Harness operations for Agent Skills: task preflight, missing-Skill discovery, local evidence, and reversible governance.
npx -y skills add CongBao/skill-steward --skill treeAssembled from the repository path, not quoted from the project. Check it against their README if it does not work.
One thing to look at
- 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.
What its author says it does
Copied from the file, not written here
Use when a task may benefit from installed or not-yet-installed Agent Skills and Skill Steward is available in the current Harness.
SKILL.md
0.9 KB, 158 tokens by cl100k_base, as published. Nobody here has run it
Skill Steward Preflight
Run skill-steward preflight --stdin --json with the exact user task on standard input.
- Prefer candidates whose decision is
use. - Present candidates whose decision is
install; never install them automatically. - Treat capability gaps as search hints, not proof that a capability is unavailable.
- Before installing a recommendation, run
skill-steward install --catalog-candidate <id> ...without--confirmand show the plan. - Add
--confirmonly after the user approves that exact plan.
Do not refresh catalogs unless the user approves the network action. Do not describe publisher classification as a safety guarantee. If Skill Steward fails, continue with the Harness normally and report the failure briefly.
What ships with it: 1 file
227 B alongside SKILL.md
agents/
- openai.yaml227 B