skill_steward

Designs public or private Agent Skill and plugin marketplaces for Cursor, Claude Code, Codex, Zed, Open Plugin, and npx skills—manifest layout, install matrix, and Skill Steward vs product boundaries. Use when setting up a marketplace, distributing skills/plugins to a team, private registry, .cursor-plugin, .codex-plugin, .plugin, .claude-plugin, or skills.sh publishing.

Establish or audit a structural quality contract for any agent-operated engineering repository: app, library, CLI/tool, plugin, harness, or meta repo. Use when a repo needs charter clarity, docs/decision ownership, type-native validation gates, evidence paths, safe action policy, cold-start legibility, maturity proof, repair-loop routing, consumer/producer owner-boundary routing, command surface evolution, evolutionary simplicity, or a repo ecology pass to decide what should be compressed, merged, removed, updated, created, retired, or moved to checks.

Scaffold and formally review a new Agent Skill in this marketplace repo. Covers valid SKILL.md generation, directory layout, registry entries, and spec auditing. Use when adding a skill, validating frontmatter, or checking marketplace readiness before a PR.

Maintains repo-local action contracts and harness repositories where product CLI and MCP adapters stay thin over core libraries. Use when adopting or improving steward.yaml actions, capability-level adoption evidence, cold-start proof loops, probes, benchmarks, CLI/MCP/core parity, adapter refactors, packages/core boundaries, or sibling harness layout; use repo-quality-system-lifecycle first for general app/library/tool stewardship baselines.

Critically analyze whether a product, repo, skill, plugin, harness, or architecture vision still fits the real application, user intent, implementation, evidence, and likely future direction. Use when asked to assess vision vs application, predict future fit, critique roadmap direction, analyze strategic intent, or decide whether to keep, refine, pivot, or kill a plan.

Maintain stewardship protocol continuity and tool/personality boundaries. Use when a repo agent needs steward-mode threshold checks, self-model update decisions, delegation hygiene, boundary violation review, or handoff-safe continuity context without overclaiming steward personality.

Replace with what this skill does and when to use it. Include trigger phrases users say, e.g. "audit my API", "scaffold a skill"; activation-critical routing belongs here.

This skill has a name that does not match its directory.

This skill incorrectly includes a README.md which should be a warning.

Name uses uppercase and underscores which violates the kebab-case rule.

A skill that is intentionally missing its license frontmatter field to test the validator warning path.

A skill that is missing its references/sources.md on purpose.

This skill is used to test the line count warning in the validator.

Does something useful when the user wants to do X. Triggered by phrases like "help me with X".

Simulates a skill directory that is well-formed but absent from skills.sh.json, to exercise registry drift detection in the validator.

This skill exists to test the body length <50 chars warning in the validator.