一键导入
skf-audit-skill
Drift detection between skill and current source code. Use when the user requests to "audit a skill" or "audit skill" for drift.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Drift detection between skill and current source code. Use when the user requests to "audit a skill" or "audit skill" for drift.
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Build data ingestion + transformation pipelines on top of cocoindex v1.0.0 — a Python framework with a Rust engine for ultra-performant, incremental indexing (ETL, RAG, knowledge graphs, vector search). Use when authoring cocoindex v1.0.0 components — App / Environment / @lifespan / @fn / mount + TargetState reconciliation + connectors + ops. **This is a complete paradigm change from v0.3.37**; the old FlowBuilder/DataScope/DataSlice/flow_def API was wholly removed in v1.0.0 — do NOT mix vocabularies.
Builds apps on top of cognee v1.0.0, the knowledge-graph memory engine for AI agents. Use when ingesting text/files/URLs into persistent memory, building knowledge graphs, searching graph-backed memory with multiple SearchType modes, enriching graphs with memify/improve, scoping memory with datasets and node_sets, configuring LLM/embedding/ graph/vector backends, running custom task pipelines, tracing operations, decorating agent entrypoints with `agent_memory`, connecting to Cognee Cloud with `serve`, or visualizing the graph. Covers cognee/__init__.py exports: the V1 API (add, cognify, search, memify, datasets, prune, update, run_custom_pipeline, config, SearchType, visualize_graph, pipelines, Drop, run_startup_migrations, tracing) and the V2 memory-oriented API (remember, RememberResult, recall, improve, forget, serve, disconnect, visualize, agent_memory). Do NOT use for: cognee internals, the HTTP REST API (use cognee-mcp or the FastAPI server), non-cognee memory/RAG libraries.
Authors Storybook v10 stories using the consolidated `storybook` package import surface for React plus Vite. Use when writing or editing `*.stories.tsx`, `preview.ts`, or `.storybook/main.ts` files on a Storybook 10.3+ project, including CSF3 story syntax, `play` functions with `storybook/test`, `preview-api` hooks, theming, MDX doc blocks from `@storybook/addon-docs/blocks`, and the a11y, themes, and vitest addons. Covers the `@storybook/react-vite` framework, `@storybook/react` renderer, and `@storybook/builder-vite` layer model so defects can be located at the right layer. Do NOT use for first-time project setup, framework selection, or upgrade migration (those flows are covered by the official upgrade CLI). Do NOT generate CSF2 default-export story arrays — v10 uses CSF3 named exports with the `satisfies` meta pattern, and training data frequently shows the wrong `@storybook/*` sub-package import paths that v10 consolidated into the single `storybook` package.
Installs React 19 and Next.js 16 UI components, blocks, and pages from the uitripled shadcn-compatible registry (the react-shadcn variant). Use when composing landing pages, dashboards, or UI from 171 animated Framer Motion components via "npx shadcn@latest add @uitripled/COMPONENT_ID". Covers the registry catalog, CLI install paths, ThemeProvider and UILibraryProvider setup, shadcn/ui primitives, and the @uitripled/utils helpers. Do NOT use this skill for the deferred react-baseui or react-carbon variants (they will ship as separate skills), and do NOT generate barrel imports from the react-shadcn package — its src/index.ts file is empty by design.
Design a skill scope through guided discovery. Use when the user requests to "create a skill brief" or "brief a skill."
Cognitive completeness verification — quality gate before export. Use when the user requests to "test a skill" or "verify skill completeness."
| name | skf-audit-skill |
| description | Drift detection between skill and current source code. Use when the user requests to "audit a skill" or "audit skill" for drift. |
Detects drift between an existing skill and its current source code, producing a severity-graded drift report with AST-backed findings and actionable remediation suggestions. Every finding must trace to actual code with file:line citations — structural truth over semantic guessing. Analysis depth adapts based on detected forge tier (Quick/Forge/Forge+/Deep) with graceful degradation. Stack skills are supported: code-mode stacks are audited per-library against their sources; compose-mode stacks check constituent freshness via metadata hash comparison.
You are a skill auditor operating in Ferris Audit mode. This is a deterministic analysis workflow — you enforce the zero-hallucination principle. You bring AST analysis expertise and drift detection methodology, while the source code provides the ground truth.
These rules apply to every step in this workflow:
stepsCompleted in output file frontmatter before loading next step{communication_language}{headless_mode} is true, auto-proceed through confirmation gates with their default action and log each auto-decision| # | Step | File | Auto-proceed |
|---|---|---|---|
| 1 | Initialize & Baseline | steps-c/step-01-init.md | No (confirm) |
| 2 | Re-Index Source | steps-c/step-02-re-index.md | Yes |
| 3 | Structural Diff | steps-c/step-03-structural-diff.md | Yes |
| 4 | Semantic Diff | steps-c/step-04-semantic-diff.md | Yes (skip at non-Deep) |
| 5 | Severity Classification | steps-c/step-05-severity-classify.md | Yes |
| 6 | Report | steps-c/step-06-report.md | Yes |
| 7 | Workflow Health Check | steps-c/step-07-health-check.md | Yes |
| Aspect | Detail |
|---|---|
| Inputs | skill_name [required] |
| Gates | step-01: Confirm Gate [C] |
| Outputs | drift-report-{timestamp}.md with drift_score and nextWorkflow |
| Headless | All gates auto-resolve with default action when {headless_mode} is true |
Load config from {project-root}/_bmad/skf/config.yaml and resolve:
project_name, output_folder, user_name, communication_language, document_output_languageskills_output_folder, forge_data_folder, sidecar_pathtimestamp as YYYYMMDD-HHmmss format. This value is fixed for the entire workflow run.Resolve {headless_mode}: true if --headless or -H was passed as an argument, or if headless_mode: true in preferences.yaml. Default: false.
Load, read the full file, and then execute ./steps-c/step-01-init.md to begin the workflow.