菜单
SkillsMP 已收集 EMPIRIC2/TAC-to-IWXXM 中的 44 个 Skill。打开任一 Skill 可查看来源和详情。
已展示 40 / 44 个已收集 Skill。
Pre-deploy gate. Verifies the deployment strategy is sound and all prerequisites are met. Walks through potential failure modes and mitigations with the user. Checks configuration, secrets, browser connectivity readiness (CORS + VITE_*), data management,…
原文语言：英语
Executes the deployment, runs API smokes (H1–H3) plus browser connectivity gates (H4–H5: CORS + frontend bundle wiring), health checks, changelog, and monitoring baseline. Final pipeline stage. Blocking — user must approve deployment results.
原文语言：英语
Orchestrator for feature and new_service sessions: adds product capabilities, scope/API/arch changes, breaking refactors, new dependencies, and multi-doc spec updates. Uses Cursor Plan mode as the default orchestrator for intake, Fn allocation, impact, and…
原文语言：英语
Reviews GitHub pull requests using the Katy Huff PR checklist, stevemao PR template checklists, deployed service CI and project rules, and Bugbot + Security Review subagents. Posts inline comments and a summary review body to the PR via gh CLI; records cycles…
原文语言：英语
Validates PyPI publish readiness for tac-validate, iwxxm-validate, and tac2iwxxm (F12–F14): OIDC trusted publishing, version tags, package boundaries, wheel copyright constraints, and hard perf gates at publish. Use before tagging, editing publish workflows,…
原文语言：英语
Investigates a deployed service in two layers: platform infra health (API up, DB migrations, secrets, deploy drift, GitHub main CI green via H0ci) and live behavior (API smokes, E2E tiers). AskQuestion-driven depth and budget before running checks.…
原文语言：英语
Recommended entry for every work session. Classifies session type (greenfield, feature, hotfix, integration, new_service, ops, process), allocates SNNN-slug, writes session-brief and routing-plan, sets active_session. Also gathers context: project mode writes…
原文语言：英语
Interviews the user to produce standing spec documents (feature-list, spec, user-journeys, test-plan, etc.) within an active session. Supports delta mode for feature/new_service sessions. If 00-context ran, pre-populates from context brief. Requires…
原文语言：英语
Verifies product plan documents by breaking them into provable statements, risk-classifying each (high/medium/low confidence), auto-approving high-confidence statements derived from user's own answers, and presenting medium/low confidence statements for user…
原文语言：英语
Creates project-specific Cursor hooks, rules, skills, and agents that prevent drift from the approved product plan. These are guardrails — not dev tools. Produces scope-checking hooks, plan-adherence rules, and domain-specific skills/agents tailored to the…
原文语言：英语
Interviews the user for technical details: architecture, deployment strategy, test plan specifics, integration points, package choices, and ADRs. Produces an execution plan, Build Plan Card (Plan-mode handoff), dependency inventory, ADRs, deployment plan, and…
原文语言：英语
Verifies technical plan documents by breaking them into provable statements, risk-classifying each, and presenting medium/low confidence statements for user review. Includes embedded consistency checking between technical plan and product plan. Runs a…
原文语言：英语
Creates development tooling: Cursor hooks for linting, formatting, typechecking, and testing; rules for spec-adherence, TDD, atomic commits, and build execution; and tool configuration files. Blocking stage — must complete before build execution begins.
原文语言：英语
Executes the implementation plan task-by-task following TDD ordering, spec-adherence rules, and atomic commit conventions. Starts each milestone batch in Cursor Plan mode (Build Plan Card) then switches to Agent mode for the Task Loop. Orchestrates parallel…
原文语言：英语
DEPRECATED redirect. Use 04-tech-plan for execution plans. Do not invoke build-planner for new work (RET-001 skill trim).
原文语言：英语
Greenfield session orchestrator: runs stages 00-context through 13-deploy-smoke for a new project inside an active_session with type greenfield. Requires 00-context to open the session and approve routing-plan.md (recommended). Combines phase gates,…
原文语言：英语
Overall verification that the implementation matches what the user wanted. Collects results from 09-qa and 10-e2e, performs feature-level completeness checking against the product plan, and walks the user through approval of each feature area. Produces…
原文语言：英语
Post-deployment surgical edits for hotfix sessions: bug fixes, patches, small behavioral changes, and dependency updates. Open via 00-context (type hotfix) or resume active_session. Use for production/local bugs — not new features (16-evolve) or greenfield…
原文语言：英语
Audits doc-planner output by breaking each document into provable (falsifiable) statements and walking the user through each one for approval, denial, or modification. Produces a document audit report, a decision log, and tracks progress with persistent state…
原文语言：英语
Executes the implementation plan from build-planner, task by task, following TDD ordering, spec-adherence rules, and atomic commit conventions. Orchestrates parallel agents for independent tasks, manages branches and PRs, runs gate checks, and keeps the…
原文语言：英语
Creates an execution plan from audited doc-planner specs, gets user approval (approve/deny/modify), sets up hooks for linting, formatting, and typechecking, and writes rules enforcing spec-adherence and TDD. Runs after doc-planner and audit-docs. Use when the…
原文语言：英语
Deploys deployed service to the configured target (Render, Docker, K8s, etc.), monitors CI/CD, validates migrations and API health, and runs post-deploy checks. Use when deploying, pushing to production, or validating a staging environment.
原文语言：英语
Plans and generates spec-driven development documents (roadmap, test plan, ADRs, feature list, API spec, etc.) for a repository. Requires the gather-context skill to have been run first — consumes the Research Brief it produces. Synthesizes findings into a…
原文语言：英语
Runs paper-analyst and repo-researcher agents in parallel, cross-references their findings, surfaces contradictions / ambiguities / decisions / bloat / uncertainty to the user via AskQuestion, and produces a Research Brief that downstream skills (doc-planner,…
原文语言：英语
Runs comprehensive verification after build-executor completes: linting, typechecking, formatting, and test suite via parallel subagents. Surfaces failures to the user with recommendations and approve/deny/modify choices. Use after build-executor passes…
原文语言：英语
DEPRECATED redirect. Use 02-verify-plan (product) or 05-verify-tech (tech). Do not invoke audit-docs for new work (RET-001 skill trim).
原文语言：英语
DEPRECATED redirect. Use 07-build for implementation. Do not invoke build-executor for new work (RET-001 skill trim).
原文语言：英语
DEPRECATED redirect. Use 13-deploy-smoke for deploy and live verification. Do not invoke deploy-verify for new work (RET-001 skill trim).
原文语言：英语
DEPRECATED redirect. Use 01-requirements for product specs. Do not invoke doc-planner for new work (RET-001 skill trim).
原文语言：英语
DEPRECATED redirect. Use 00-context for session open and context briefs. Do not invoke gather-context for new work (RET-001 skill trim).
原文语言：英语
DEPRECATED redirect. Use 08-verify-build for milestone gates. Do not invoke verify-build for new work (RET-001 skill trim).
原文语言：英语
Download or copy a PDF into the repo, extract page-marked text under .local/ (gitignored), and write focused mining notes under docs/domain/mining/ (transitory). Promote durable citations into TAC_VALIDATION / IWXXM_CONVERSION / IWXXM_VALIDATION via…
原文语言：英语
Mines authoritative TAC validation, TAC→IWXXM conversion, and IWXXM validation sources into docs/domain/: lean canonicals (TAC_VALIDATION, IWXXM_CONVERSION, IWXXM_VALIDATION) + rules/ catalog, with dense digs under mining/ (transitory). Use when the user asks…
原文语言：英语
Canonical workflow when any failure is caught (14-hotfix, 15-service-health, build, e2e, CI). One bug = one markdown report in docs/bug-reports/, one repro test in tests/bugs/, one fix. Requires error logs + description, persistent investigation in the MD…
原文语言：英语
Runs quality checks at milestone boundaries during the build. Auto-corrects lint and format issues without blocking. Only blocks on test failures, type errors, or security issues that require user decisions. Invoked by 07-build at milestone and phase…
原文语言：英语
Post-build quality assurance. Runs full lint, format, typecheck, security, dependency, data-staging, Modal workspace, and cross-file checks on the entire codebase. Writes `{artifacts_dir}/reports/qa-report.md` with blocking vs advisory findings. Runs in…
原文语言：英语
Verifies that user journeys from product requirements work end-to-end. Mechanism is adaptive: browser automation for web apps, API calls for services, shell execution for CLIs, import/call for libraries. Runs asynchronously in parallel with 09-qa. Results…
原文语言：英语
Reviews Cursor agent conversation logs and pipeline skills 00-19 against project artifacts, then interviews the user with batched AskQuestion prompts to capture what went well, what to improve, and brainstorm process fixes. Ends with an interactive…
原文语言：英语
Works through pull request review findings one-by-one or in batches: triages GitHub review comments (including 18-pr-review output), confirms fix approach with the user, applies TDD for blockers, commits atomically, replies and resolves threads, pushes,…
原文语言：英语
Add and update NumPy-style docstrings for Python modules, classes, and functions, and TSDoc comments for TypeScript modules, classes, interfaces, and functions. Use when documenting code, writing docstrings, adding API comments, improving module docs, or when…
原文语言：英语