スキルを探す

On-demand "what's new" generator. Main context (which knows the impending change) writes keepachangelog-style entries to a fragment file at `.claude/state/whatsnew/<slug>.json` (gitignored, transient). An optional per-project `project.json → whatsnew.route_workflow` names a routing workflow that consumes the fragment and renders it wherever the project wants (a site page, a release body, a docs page). Not a workflow phase — it never blocks a commit, and it never writes `CHANGELOG.md` (that file is owned solely by semantic-release in CI).

日本語の概要は準備中です。原文の説明を表示しています。

Execute a swarm plan wave by wave with filesystem isolation via git worktrees. For each wave, main context decides the scenario recipe + implementation contract for every task, then spawns one swarm-worker per task in parallel. Each worker executes its recipe and reports JSON status. Worktree merge-audit verifies write-set discipline before changes land on main. Aborts remaining waves on any audit or task failure.

日本語の概要は準備中です。原文の説明を表示しています。

Triage an incoming request — pick the workflow entry phase (intake / spec / tdd / chore) and record the workflow statefile that the Track Guard reads.

日本語の概要は準備中です。原文の説明を表示しています。

Decompose an approved spec into a dependency-ordered swarm plan — one task per component, each with an explicit write_set. Produces `.claude/state/swarm/<slug>.json` with tasks + waves. The wave scheduler guarantees pairwise-disjoint write_sets within each wave so parallel dispatch is provably conflict-free.

日本語の概要は準備中です。原文の説明を表示しています。

Workflow Phase 6 — TDD coordinator. Decides the scenario recipe and the implementation contract in main context, writes them to a state file, seeds per-worker tasks (scenario, implement, verify-tick, design-ui-tick) into the TaskList, and yields with harness_state continue so the harness invokes each worker as its own tick. No subagent delegation; no nested Skill calls.

日本語の概要は準備中です。原文の説明を表示しています。

Read-only release + backlog recap. Reports the last release, commits-since-tag classified by conventional-commit type with the semver bump they trigger and pushed-vs-origin state, the backlog bucketed (open/picked-up/dropped with epic parent→child nesting), the roadmap epics with their per-task tallies, and condensed open questions — then recommends the next pickup. One `cli.mjs recap` call returns all six recap keys. Invoke any time (on demand) to plan a release or choose the next thing to build; a compact form is also surfaced at session start. Not a workflow phase; never writes CHANGELOG; never starts or commits work.

日本語の概要は準備中です。原文の説明を表示しています。

Report the health of the central system spec at `docs/system/` across seven checks — coverage gaps, stale elements, dangling anchors, duplicate anchors, orphan shards, unillustrated elements, and shards missing their kind annotation. Report-first and read-only: it repairs nothing until a human confirms a specific item. Use when the corpus has drifted, after a merge that touched `docs/system/`, or from `/archive` Step 5.5 in report-only mode. Requires `memory.architecture_map.enabled`.

日本語の概要は準備中です。原文の説明を表示しています。

Write technical documentation that matches the measured profile of professionally-written docs. Carries the craft rules derived from 114k words of pre-2022 documentation (SQLite, PostgreSQL, Python, Django, Rust, Go, Pro Git, nginx), the per-Diátaxis-type numeric targets, and the copywriting register for page leads. Use when drafting or rewriting any documentation prose, or when a page is factually right but reads as machine-written.

日本語の概要は準備中です。原文の説明を表示しています。

Reconcile baseline-versioned files that the `create-baseline upgrade` CLI staged for LLM-assisted semantic merge. Use when the CLI prints "Open Claude Code and run /upgrade-project to reconcile". Reads the stage manifest at `.claude/state/upgrade/<ts>/manifest.json`, reasons through each pending file's three-way delta in main context, writes a reconciled LOCAL, then deletes the stage when every file lands. Supports `--dry-run` (preview the reconciled diff without writing) and a structured "needs-user-input" fallback when the conflict cannot be disambiguated automatically.

日本語の概要は準備中です。原文の説明を表示しています。

When the user wants to create step-by-step technical tutorials, quickstarts, or code walkthroughs. Trigger phrases include "tutorial," "quickstart," "getting started guide," "walkthrough," "step by step," "how to guide," "hands-on guide," or "code tutorial."

日本語の概要は準備中です。原文の説明を表示しています。

Contract document for the binding test verdict file at .claude/state/last_test_result. Format spec for callers (integrate, simplify, chore, the verify-tick worker) that inline the four mechanical operations. Not Skill-tool-invocable.

日本語の概要は準備中です。原文の説明を表示しています。

Standard operating procedure for producing a documentation page end to end — gather source material, classify, draft against the measured corpus profile, then run the reader-level and humanizer passes and both gates. Use for any page on a documentation surface, or when rewriting a page that reads as machine-written. Orchestrates technical-writing, reader-level and humanizer in a fixed order.

日本語の概要は準備中です。原文の説明を表示しています。

Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, negative parallelisms, and excessive conjunctive phrases. Credits: Original skill by @blader - https://github.com/blader/humanizer

日本語の概要は準備中です。原文の説明を表示しています。

Capture what someone actually said without contaminating it with your own inference. Invoke when eliciting intent in conversation (design discussion, requirement interview, decision surfacing) and when turning a conversation into a durable record (ADR, decision node, memory entry, spec `## Decisions` row). Enforces quote-before-critique, keeps illustration separate from commitment, and holds interpretation subordinate to verbatim. Prevents the failure where unspecified space is filled by the recorder and then evaluated as if the speaker had specified it.

日本語の概要は準備中です。原文の説明を表示しています。

MANDATORY skill for ALL code generation. Enforces top-down composition, consistent abstraction layers, and proper module hierarchy. Apply every time you write or modify code — in any language — no exceptions.

日本語の概要は準備中です。原文の説明を表示しています。

Write and maintain technical documentation. Trigger with "write docs for", "document this", "create a README", "write a runbook", "onboarding guide", or when the user needs help with any form of technical writing — API docs, architecture docs, or operational runbooks.

日本語の概要は準備中です。原文の説明を表示しています。

Workflow Phase 10 — orchestrator for documentation work. Surveys the diff, routes any page on a documentation surface through `technical-writer`, standalone technical reference through the `documentation` skill, tutorials through `technical-tutorials`, and all other prose body work (inline docs, README updates, user-facing copy) through the `prose` skill (which mandates `humanizer` and conditionally `copywriting`). Verifies completeness and marks phase done.

日本語の概要は準備中です。原文の説明を表示しています。

The primary tool for code-navigation questions in ANY language (frontend or backend) — walk the structural graph from an entry point (page, route, handler, command, `main`) down through imports to the IO boundary (network call, query, file read). Use this skill, in preference to the Explore agent or global grep, whenever the user asks "where does X come from?", "what API/data populates Y?", "what renders or wraps the Z panel?", "which file for feature F?", "how is this wired up?", or any code-navigation question mapping a UI element, route, or feature concept to source files. Also reverse queries ("what uses /api/foo?", "who consumes this service?"). Keyword search routinely picks up unrelated flows sharing a domain word and produces wrong answers; the walk follows the actual graph. The JS/TS `walk.mjs`/`discover.mjs` are an optional accelerator, not required — the walk is language-agnostic.

日本語の概要は準備中です。原文の説明を表示しています。

Draft a Business Requirements Document (BRD) for cross-functional or stakeholder-heavy work that needs more structure than an intake. Use after `/intake` when the request spans multiple systems/teams, carries regulatory weight, or needs formal sign-off. Output lives at `docs/brd/<slug>.md`.

日本語の概要は準備中です。原文の説明を表示しています。

Draft a Root Cause Analysis for an incident, outage, or repeated test failure. Unlike intake/spec/brd, RCA is not a workflow phase — it's a standalone postmortem artifact that often precedes a bugfix intake. Output lives at `docs/rca/<slug>.md`.

日本語の概要は準備中です。原文の説明を表示しています。

Draft or revise English prose for any brief — documentation body, intake problem statements, spec context, RCA summaries, marketing copy, README sections, PR descriptions. Mandatorily invokes `Skill(humanizer)` as the final pass on every draft. Conditionally invokes `copywriting` for persuasive register, `documentation` for reference docs, `technical-tutorials` for tutorials. Used when any phase needs human-readable prose written or rewritten. Composition only — research and register-picking happen in the caller's context.

日本語の概要は準備中です。原文の説明を表示しています。

Orchestrates `impeccable` for every design task inside a workflow phase. Captures intent in natural language, classifies it (design / development / copy), translates design intents into a sequence of impeccable subcommand invocations, runs them in main context with state persistence at `.claude/state/design/<slug>.json`, and returns a structured report. ALWAYS invokes `impeccable` under the hood for the underlying design move; never picks aesthetic direction itself; never writes product code directly. Per CLAUDE.md Article X.2, all design tasks in workflow phases route through this skill.

日本語の概要は準備中です。原文の説明を表示しています。

Workflow Phase 3 — Research and Solution Exploration. Surfaces 2–4 candidate solution approaches with concrete tradeoffs, grounded in current library docs (fetched through the provider named in `.claude/docs-provider.json`) — never in training-data recall. Output lives at `docs/research/<slug>.md`. Candidate ranking and the memo execute in main context; doc/source gathering MAY be delegated to read-only advisory subagents (seed.md §4.2-A).

日本語の概要は準備中です。原文の説明を表示しています。

Power batch-sprint helper — hosts the two behaviours that distinguish the `power` workflow track from spec-entry. (1) Per-ticket iteration: security runs once PER TICKET over workflow.json.tickets[] while the mechanical phases run once for the batch. (2) Commit split: at the commit phase, group the batch's working tree into ordered Conventional Commits via commit-split.mjs, closure last. Invoked within a `power`-track workflow; not a standalone feature. Opt-in via velocity.power_mode.enabled; requires git.

日本語の概要は準備中です。原文の説明を表示しています。