# Roll Skill Catalog

> GENERATED by `roll setup skills` — do not edit by hand.
> 由 `roll setup skills` 生成 — 请勿手工编辑。
>
> Source of truth: each skill's `skills/<name>/SKILL.md` frontmatter.
> 事实源：各 skill 的 `skills/<name>/SKILL.md` frontmatter。

| Skill | Description |
|-------|-------------|
| `roll-.changelog` | Load when a completed Roll build or deploy needs CHANGELOG.md generated or updated from Done stories, acceptance evidence, and release notes. |
| `roll-.clarify` | Load when roll-build Fly mode input is vague or underspecified and the agent must ask targeted scope questions before planning or coding. |
| `roll-.dream` | Load when an architecture health scan should inspect code structure, dead code, doc staleness, and refactor candidates. Invoked on demand via roll dream run-once. |
| `roll-.echo` | Load when user intent is vague, contradictory, rambling, or unclear and should be restated for confirmation before action. |
| `roll-.qa` | Load when a build, fix, review, or story needs Roll QA coverage standards: test pyramid, evidence expectations, visual/E2E boundaries, or CI gates. |
| `roll-.review` | Load when a TCR micro-step needs self code review before commit, focused on bugs, regressions, security, and design issues. |
| `roll-build` | Load when a user gives a US-XXX story or asks to ship a feature through Roll TCR delivery, from clarification through verification and backlog write-back. |
| `roll-debug` | Load when a web page needs black-box browser diagnostics, console/network/state capture, root-cause analysis, and source fixes for project-owned issues. |
| `roll-delta-team` | Load when the current coding-agent main session must run a host-guided Delta Team delivery: it acts as implicit Supervisor and drives a Designer → Builder → Evaluator artifact-only handoff through the roll delta CLI, requesting and attesting D/B/E sub-agents with its own host facilities — Roll never spawns a session. |
| `roll-design` | Load when the user wants to discuss approaches, design a solution, model domains, split work into INVEST stories, or write backlog/spec artifacts without coding. |
| `roll-doc-audit` | Load when checking README, guides, site pages, CLI help, and docs against implemented behavior, or when auditing documentation inventory, docs/INDEX coverage, undocumented modules, and draft fills from code evidence. |
| `roll-doctor` | Load when diagnosing Roll toolchain health, conventions sync, skill file health, templates, symlinks, config validity, or installation integrity. |
| `roll-fix` | Load when the user provides a FIX-XXX or BUG-XXX item, or asks for a focused hotfix/bugfix through Roll lighter TCR repair workflow. |
| `roll-idea` | Load when the user gives a short idea or bug note that should be quickly classified, assigned an ID, and appended to backlog. |
| `roll-loop` | Load when configuring, explaining, or operating Roll session-driven backlog execution loop that takes Todo work and dispatches US/FIX/REFACTOR items. |
| `roll-notes` | Load when recording a development moment, decision, discovery, success, or failure into Roll chronological project diary. |
| `roll-onboard` | Load when bringing an existing codebase without Roll markers into Roll through read-only diagnosis and structured .roll/init-diagnosis.yaml + .roll/onboard-plan.yaml artifacts. |
| `roll-peer` | Load when the user explicitly asks for peer review, cross-agent negotiation, /peer, or a high-risk decision needs an external agent perspective. |
| `roll-prime` | Load when coordinating a Roll project as Supervisor (supervise role): reconcile backlog truth, advise the owner, dispatch the Delta Team, watch cycles read-only, diagnose structural failures, and reconcile .roll meta — not when implementing a Story as Builder. |
| `roll-propose` | Load when the owner asks for product proposal drafts from project context that should go to .roll/proposals.md, not directly into backlog. |
| `roll-review-pr` | Load when reviewing a pull request diff and emitting APPROVE, REQUEST_CHANGES, or UNCERTAIN with file/line-grounded findings. |
| `roll-spar` | Load when high-risk logic needs adversarial TDD with attacker tests and defender implementation, such as auth, payments, data integrity, or state machines. |