메뉴
Use when revising one or more existing Claude Code skills, aligning skill instructions with current repository policy, narrowing overlapping ownership, or applying a deliberate behavior change without creating a competing skill.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Creates a new specialist agent as two stitched source files, base.md plus frontmatter/claude.json, proposing model, effort, and permissions by role archetype and confirming them with the user before writing. Use when adding a new subagent, defining a new specialist role, scaffolding an agent definition, or when update-agent hands off new-agent creation.
Discover, validate, stitch, and install specialist agent templates contributed by Essential and other enabled plugins in the same marketplace. Use when asked to install agents, set up subagents, refresh the agent team, or configure Claude Code on a new machine.
[Description] — third person, what the skill does plus a 'Use when...' clause with concrete trigger phrases a user would actually type; 25-60 words; lead with the key use case.
Update explicitly selected agent definitions to the current two-file template or a stated behavior change while preserving useful role expertise, trigger boundaries, context, collaboration links, and working voice. Use when migrating agents to a template revision, correcting agent configuration, or batch-updating selected agents; require an exact selector and route genuinely new roles to create-agent.
Enforce coding standards mechanically across a selected scope with batched linters and independent reviewers. Use when source files need lint-error correction, standards enforcement, or consistent formatting, including calls extended by another plugin's portable lint profile; behavior-changing repairs belong to fix.
Review semantic correctness, security, test intent, documentation, sibling consistency, and alignment with the implementation plan. Use after code changes or for explicit review requests; report findings without editing code and leave mechanical standards enforcement to lint.
| name | update-skill |
| description | Use when revising one or more existing Claude Code skills, aligning skill instructions with current repository policy, narrowing overlapping ownership, or applying a deliberate behavior change without creating a competing skill. |
| model | opus |
| context | fork |
| allowed-tools | Bash, Task, Read, Glob, Edit, MultiEdit, TodoWrite |
| argument-hint | [skill specifier] [--changes=...] |
create-skill when no
suitable owner exists.Follow ${CLAUDE_SKILL_DIR}/../../constitution/references/authoring-invariants.md.
The current skill and its real callers are authoritative; the template is a
concise aid, not a migration target whose headings must be copied.
--all.${TMPDIR:-/tmp}/check.md) using
${CLAUDE_SKILL_DIR}/../../constitution/references/check.md as the
example table format with :white_check_mark:/:x: status markers,
and delete them before staging; they are not deliverables and must not be
committed.Independent targets may be delegated in bounded batches — at most 8 skills per
batch and 8 parallel Task calls per dispatch. Each assignment must name exact
paths and constraints; review the combined diff afterward.
claude plugin validate --strict <plugin-path>
python3 "${CLAUDE_SKILL_DIR}/../verify-skill/scripts/quick_validate.py" <target>
Use governance:verify-skill for functional and trigger evaluation with
fix: true; in bulk updates bound the loop to 2 fix iterations per skill,
record remaining issues, and continue to the next target.
Report updated skills, ownership changes, validation results, thought-experiment and blindspot coverage, runtime evaluation status (including "not exercised"), and unresolved ambiguity. Confirm temporary Markdown thought-experiment notes were deleted before commit. Never claim a bulk update without listing its targets.