Menu
Generate compact AI-readable context maps from codebases — tools like codesight, repomix, agentic-context that pre-compute project structure to save tokens in AI coding sessions.
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
Run AI coding agents on disposable repo clones. The agent works on a clone — it can't touch your real repo. An optional container (Podman or Docker) provides build/test isolation. You review the diff and decide what (if anything) to apply.
Create Deezer playlists programmatically from any query — similar artists, genre mixes, festival lineups, mood-based collections. Four-tier data pipeline: Deezer public REST API + Last.fm scrobble data for discovery, GQL Pipe API for smart mixes and playlist creation, web search for subjective curation. Uses ARL cookie auth — no OAuth app required.
Research-focused query handling with multi-source synthesis, citations, and Obsidian persistence. Like a self-hosted Perplexity/Vane but CLI-native. Best for quick-to-medium lookups using Kagi. Use when the user asks factual questions, needs citations, or wants a direct answer — not a full research report (use deep-research) or social sentiment (use last30days). Triggers on: research, look into, what's the latest on, compare, explain, investigate.
Write articles, guides, blog posts, tutorials, newsletter issues, research reports, and deep research outputs in a distinctive voice derived from supplied examples or brand guidance. Use when the user wants polished written content longer than a paragraph, deep research on a topic, or a research report — especially when voice consistency, structure, and credibility matter. Triggers on: 'write an article', 'research report', 'deep research', 'long-form', 'blog post', 'guide', 'newsletter', 'white paper', 'research paper'.
Systematic research methodology for major consumer durables (appliances, HVAC, power tools, outdoor equipment) and smart garden/outdoor devices (bird feeder cameras, bird baths, smart outdoor gadgets). Emphasis on real reliability data, failure mode analysis, and head-to-head comparison. Use when the user asks to research, review, compare, or evaluate major purchases where longevity and repair risk matter, or when researching smart bird feeders, bird bath cameras, and similar connected outdoor devices. Triggers on: washer/dryer, refrigerator, dishwasher, HVAC, furnace, AC, generator, power tool, appliance reviews, appliance reliability, which [appliance] to buy, compare models, bird feeder camera, bird bath camera, smart garden devices.
Research and evaluate gray-market / cross-border products not officially sold in the user's region. Covers import smartphone and electronics purchasing: brand availability, carrier band compatibility checks, Global vs. regional ROM verification, reputable gray-market retailers, financing options, warranty realities, and pricing benchmarks. Also covers website legitimacy evaluation — vetting unfamiliar e-commerce sites for scam risk before purchase. Triggers on: import phone, buy [Xiaomi/Oppo/Vivo/Realme/Redmi] in US, gray market, international phone, will this phone work on [carrier], where to buy [product] not sold here, import electronics, is [website] legit, is [website] a scam, can I trust [website], open-ear earbuds alternatives, import earbuds, Huawei FreeClip, buy [product] outside [region].
| name | ai-context-generation |
| description | Generate compact AI-readable context maps from codebases — tools like codesight, repomix, agentic-context that pre-compute project structure to save tokens in AI coding sessions. |
| version | 1.0.0 |
| author | Hermes Agent |
| license | MIT |
| platforms | ["linux","macos","windows"] |
| metadata | {"hermes":{"tags":["context-engineering","token-savings","codebase-analysis","AI-coding","codesight"]}} |
Tools that scan a codebase and produce compact, AI-readable context maps (markdown, wiki, or MCP tool sets) so AI coding assistants understand the project without burning tokens on file exploration.
npx codesight — Universal AI context generator. Zero dependencies, MIT, TypeScript. 1.1K+ stars.
Strengths:
--blast <file> shows affected files at 3 hops depth--mcp)--init produces CLAUDE.md, .cursorrules, codex.md, AGENTS.md--mode knowledge <path> maps markdown notes (Obsidian vaults, ADRs) into KNOWLEDGE.md--wiki creates a .codesight/wiki/ knowledge baseLimitations (web/TS-centric):
Commands:
npx codesight # Basic scan → .codesight/CODESIGHT.md
npx codesight --blast <file> # Blast radius: what breaks if you change a file
npx codesight --benchmark # Token savings breakdown
npx codesight --init # Generate CLAUDE.md, .cursorrules, codex.md, AGENTS.md
npx codesight --mcp # Run as MCP server (13 tools)
npx codesight --wiki # Generate wiki knowledge base
npx codesight --mode knowledge <path> # Map markdown notes → KNOWLEDGE.md
npx codesight --profile <tool> # Generate optimized config for a specific AI tool
npx codesight --open # Open interactive HTML report in browser
Output files (.codesight/):
CODESIGHT.md — combined context map (one file, full project understanding)routes.md — API routes with method, path, params (web only)schema.md — DB models with fields, types, keys, relations (web only)components.md — UI components with props (web only)libs.md — library exports with function signatures (web only)config.md — env vars, config files, key deps (web only)middleware.md — auth, CORS, rate limiting, etc. (web only)graph.md — import relationships and hot files (language-agnostic)report.html — interactive visual dashboard (with --html or --open)npx codesight --benchmark first to see what detectors fired. If routes/models/components are all 0, the project type isn't supported for structural extraction..codesight/ if not integrating into workflow — the directory is not gitignored by default. Decide whether to commit it or remove it.codebase-inspection skill (pygount) — LOC counting and language breakdown, complementary but different purpose