Loading...
Loading...
Found 43 Skills
Create, review, and update Prompt and agents and workflows. Covers 5 workflow patterns, agent delegation, Handoffs, Context Engineering. Use for any .agent.md file work or multi-agent system design. Triggers on 'agent workflow', 'create agent', 'ワークフロー設計'.
Internal step of plan-feature: turn a feature-request issue into a scoped, sized, roadmap-mapped SPEC **product half** (capability closure satisfied) with Closes #N traceability.
Read-only workflow sensor: compute repository, roadmap, dependency, PR, finding, and recovery state, then emit the fixed machine envelope. Never edits. Triggers: "workflow-status", "workflow status", "what can I build next", "state of the run".
Review a change with only applicable internal axes, classify every finding, persist fix-now work, and return one evidence-backed decision. Findings only; --adversarial N uses isolated reviewers; --synthesize fuses supplied reviewer tables. Triggers: "review-change", "review this change", "adversarial review".
Internal contract: one compact frozen ACCEPTANCE.md per delivery unit, its validation ladder, anti-weakening rules, and blob-bound execution receipt. Consumed by planners, execute-phase, review-change, and loop-review-fold.
Internal contract: the single owner of the eight phase-lint rules, the fixed PASS/BLOCKED result, and the normalized phase fingerprint. Consumed by plan-feature-scaffold, plan-fix, and execute-phase. Not a menu entry.
Internal scope/classification engine composed by review-change (and reused by the audit skills): consumes the synthesized findings table, verifies every applicable axis is represented, and classifies each finding into a decision table (fix-now / replan-in-unit / decision-required / proposal / ignore). Findings only — never refactors.
Internal planning gate: consumes the normalized repository state and makes the ONE final architectural classification, using a two-stage contract that allows a lightweight NRS read during planning and a full architectural classification after the plan is cut. Used by plan-feature, plan-feature-from-issue, plan-feature-scaffold, and plan-fix. Not a menu entry.
Use when lasting learnings from the current session should be recorded in agent instruction files - AGENTS.md, CLAUDE.md, or the user's global instructions - such as pitfalls hit, mistakes corrected, conventions found, or commands that finally worked, whatever the conversation language. Proposes minimal, deduplicated edits per file and applies only the approved ones. Not for general memory, a full audit of instruction files, or improvement requests for skills.
You MUST use this when verified work awaits a decision on how it leaves the branch - merging it, pushing it, opening a pull request (PR), keeping the branch, discarding the branch, or cleaning up a worktree or workspace - in any repository. Not for work that is not verified yet (verify), open review findings (cr-received), a plain commit with no integration decision, or release work with its own checklist. Replaces devflow branch-finish.
Use when code review feedback has arrived and needs a decision before any fix - a cr-request result, pull request comments, CI or static-analysis review output, or a reviewer's message - in any programming language. Verifies each finding against the current code, decides accept, reject, partial, defer, or escalate with evidence, fixes only what holds up, and reports per finding.
Use when a change needs an independent code review before it moves on - a finished task, a feature branch, a pull request, or work-in-progress changes since a given ref - in any programming language. Pins the exact diff, gathers intent and constraints, and dispatches a read-only reviewer that returns separate verdicts and evidence-backed findings.