{
"markdown": "# vibe-coding-mcp\n\n> Auto-documents your AI coding sessions — wire it as a Claude Code hook and your daily README / design-decision log / refactor-PR description writes itself.\n\n[](https://github.com/MUSE-CODE-SPACE/vibe-coding-mcp/actions/workflows/ci.yml)\n[](https://www.npmjs.com/package/vibe-coding-mcp)\n[](https://opensource.org/licenses/MIT)\n\n\n[English](#english) · [한국어](#한국어)\n\n---\n\n\n## Why vibe-coding?\n\nYou spent four hours pair-programming with Claude. You made three architectural decisions, wrote eight files, and reverted twice. Tomorrow you will not remember why you picked the second option in commit `a1b2c3d` over the first one you tried. Most people solve this with manual journaling that lasts two weeks, or by trawling `git log` after the fact.\n\nvibe-coding-mcp solves it differently: it sits as an MCP server next to your editor, **collects code blocks and design decisions out of the conversation**, generates **README / DESIGN / API / ARCHITECTURE** documents from them, and **publishes to Notion / GitHub Wiki / Obsidian / Confluence / Slack / Discord** — all from inside the chat with Claude. Wire it as a Claude Code `PostToolUse` + `Stop` hook (one block of JSON, see [Quickstart](#5-minute-quickstart)) and every session auto-captures into `~/.vibe-coding-mcp/sessions/` without you typing anything. The bundled `daily-vibe-log` prompt then rolls today's captures into one document.\n\nThe Phase 3 refactor (v2.14.0, May 2026) collapsed the two-entry-point drift (`index.ts` had 7 tools, `stdio.ts` had 15) into a **single transport-agnostic registry**, so HTTP and stdio now expose the same 15 tools, 3 resources, and 3 prompts.\n\n## What's new in v2.14.0\n\n- **One registry, two entry points, zero drift.** `src/core/toolRegistry.ts` is the single source of truth — both `index.ts` (Streamable HTTP) and `stdio.ts` (bin) load the same 15 tools, 3 resources, and 3 prompts.\n- **Migrated to high-level `McpServer` API** (SDK 1.25+). `McpServer.registerTool() / .resource() / .prompt()` instead of low-level `setRequestHandler(CallToolRequestSchema)`. Adding a tool is now a one-line `register()` call.\n- **Streamable HTTP (MCP 2025-03-26), not SSE.** `POST /mcp` at `:3000`. The legacy `/sse` route returns HTTP 410 with a pointer.\n- **3 new MCP Resources for `@-mention`.** `vibe-coding://sessions/list` (captured sessions), `vibe-coding://sessions/{id}` (one session in full), `vibe-coding://config` (current platform creds).\n- **3 new MCP Prompts.** `daily-vibe-log` (roll today's sessions into one doc), `document-session` (one session → README/DESIGN/API/etc. → publish), `refactor-context` (session → PR description with decisions + AST analysis + git diff).\n- **+34 tests** (149 → 183 passing) covering the registry, resources, prompts, and the McpServer factory.\n\n## 5-minute Quickstart\n\n```bash\n# 1. Install via Claude Code\nclaude mcp add vibe-coding-mcp -- npx -y vibe-coding-mcp\n```\n\n2. Wire it as a Claude Code hook so every session auto-captures. Add this to `~/.claude/settings.json`:\n\n```jsonc\n{\n \"mcpServers\": {\n \"vibe-coding-mcp\": {\n \"command\": \"npx\",\n \"args\": [\"-y\", \"vibe-coding-mcp\"]\n }\n },\n \"hooks\": {\n \"PostToolUse\": [\n {\n \"matcher\": \"Edit|Write|NotebookEdit\",\n \"hooks\": [{\n \"type\": \"command\",\n \"command\": \"claude mcp call vibe-coding-mcp muse_session_history '{\\\"action\\\":\\\"save\\\",\\\"title\\\":\\\"auto-capture\\\",\\\"summary\\\":\\\"PostToolUse\\\",\\\"tags\\\":[\\\"auto-capture\\\"]}' >/dev/null 2>&1 || true\"\n }]\n }\n ],\n \"Stop\": [\n {\n \"hooks\": [{\n \"type\": \"command\",\n \"command\": \"claude mcp call vibe-coding-mcp muse_create_session_log '{\\\"title\\\":\\\"Claude Code session\\\",\\\"summary\\\":\\\"Stop hook\\\",\\\"options\\\":{\\\"logType\\\":\\\"session\\\"}}' >/dev/null 2>&1 || true\"\n }]\n }\n ]\n }\n}\n```\n\n3. Restart Claude Code. The next time you edit a file in a session, capture starts.\n4. After a session, try this in Claude:\n\n```\n오늘 캡처된 세션들 합쳐서 daily vibe log 만들어줘. (use the /daily-vibe-log prompt)\n```\n\n5. Expected: Claude calls `muse_session_history(action='list', filterTags=['auto-capture'])` to gather today's sessions, then `muse_create_session_log` to compose, then offers to publish to Notion / Obsidian / GitHub Wiki. See [`docs/AUTO_CAPTURE.md`](./docs/AUTO_CAPTURE.md) for hook customization (e.g. piping `git diff --stat` into the capture payload).\n\n## Real use cases\n\n### 1. \"I want today's coding session as a daily log without lifting a finger\"\n**Problem:** You'll never sit down at the end of the day and write a journal entry. Nobody does.\n**With this MCP:** the `PostToolUse` + `Stop` hooks save sessions automatically into `~/.vibe-coding-mcp/sessions/`. Next morning, run the `/daily-vibe-log` prompt and Claude reads `vibe-coding://sessions/list` for today, compiles a Markdown log (problems, decisions, code blocks, blockers), and offers to publish to Notion.\n**Why it's better than git log or Notion templates:** git log knows *what* you committed, not *why* you picked option B over option A. The session capture preserves the conversation context that the diff doesn't.\n\n### 2. \"I need a PR description for a refactor that touched 11 files\"\n**Problem:** Writing a good PR description for a refactor takes 15 minutes of re-reading the diff. Most refactor PRs end up with a one-liner and reviewers hate it.\n**With this MCP:** invoke the `/refactor-context` prompt against the session you just finished. The prompt chains `muse_session_history(get)` → `muse_summarize_design_decisions` → `muse_analyze_code` (AST + Mermaid diagram) → `muse_git(diff)` and outputs a structured PR description with: *Why* (decisions), *What changed* (file-by-file from AST), *Trade-offs* (decisions section), and *How to review* (diagram of new architecture).\n**Why it's better than asking Claude to \"write a PR description\":** the prompt explicitly pulls the **design-decision summary** out of the session, not just the diff — reviewers get the why, not a paraphrased diff.\n\n### 3. \"I want my Obsidian vault to fill itself with my design decisions\"\n**Problem:** You make architectural decisions during pair-programming and they evaporate. ADR templates require manual discipline you don't have.\n**With this MCP:** call `muse_summarize_design_decisions` over your session, then `muse_publish_document({ platform: 'obsidian', vault: '~/MyVault/decisions' })`. The Obsidian platform writer adds YAML frontmatter (tags, date, session-id back-reference), so the new note shows up correctly in your vault graph view.\n**Why it's better than a `Notes/ADR-template.md`:** the decision text is generated from the actual code+conversation, not from your tired post-session memory. And it works across **6 platforms** (Notion, GitHub Wiki, Obsidian, Confluence, Slack, Discord) with one tool call.\n\n## Tools / Resources / Prompts\n\n| Name | Type | What it does |\n|------|------|--------------|\n| `muse_collect_code_context` | tool | Pull code blocks + conversation summary out of a chat into a structured session |\n| `muse_summarize_design_decisions` | tool | Extract architectural / design decisions (problem → options → choice → trade-off) |\n| `muse_analyze_code` | tool | AST analysis (TypeScript / Python / Go) + Mermaid diagrams (Class / Flow / Sequence / ER / Architecture) |\n| `muse_generate_dev_document` | tool | README / DESIGN / TUTORIAL / CHANGELOG / API / ARCHITECTURE generator |\n| `muse_normalize_for_platform` | tool | Markdown normalization for Notion / GitHub Wiki / Obsidian / Confluence / Slack / Discord quirks |\n| `muse_publish_document` | tool | Publish to any of the 6 supported platforms |\n| `muse_create_session_log` | tool | Daily or per-session log composition |\n| `muse_session_history` | tool | `save / get / update / delete / list / search / stats` over `~/.vibe-coding-mcp/sessions/` |\n| `muse_export_session` | tool | Export one session to Markdown / JSON / HTML |\n| `muse_project_profile` | tool | Per-project settings (default platform, default tags, language) |\n| `muse_git` | tool | `status / log / diff / branch / snapshot` + design-decision extraction from commit messages |\n| `muse_session_stats` | tool | Productivity dashboard: sessions/day, decisions/session, language breakdown |\n| `muse_auto_tag` | tool | AI tag suggestions for a session (Claude API, optional) |\n| `muse_template` | tool | Custom doc templates (per project / per output type) |\n| `muse_batch` | tool | Compose multiple tool calls sequentially or in parallel in one round-trip |\n| `vibe-coding://sessions/list` | resource | List of captured sessions (`@-mention`-able) |\n| `vibe-coding://sessions/{id}` | resource | One session's full body, code blocks, decisions, tags |\n| `vibe-coding://config` | resource | Current platform configuration (which integrations are wired) |\n| `prompt://vibe-coding/daily-vibe-log` | prompt | Roll today's captured sessions into one daily log |\n| `prompt://vibe-coding/document-session` | prompt | One session → dev document → publish |\n| `prompt://vibe-coding/refactor-context` | prompt | Session → PR description with decisions + AST + git diff |\n\n## How it works\n\n```\n Claude (or any MCP client)\n │\n ┌────────────────────────┴────────────────────────┐\n ▼ ▼\n src/index.ts (Streamable HTTP, POST /mcp at :3000) src/stdio.ts (bin)\n │ │\n └────────────────────────┬─────────────────────────┘\n ▼\n src/core/mcpServerFactory.ts\n │\n ┌─────────────────┼─────────────────┐\n ▼ ▼ ▼\n toolRegistry.ts resources.ts prompts.ts\n (15 muse_* tools) (3 resources) (3 prompts)\n │\n ▼\n src/tools/*.ts (15 tool implementations)\n │\n ▼\n src/core/sessionStorage.ts (~/.vibe-coding-mcp/sessions/)\n src/platforms/*.ts (notion / github-wiki / obsidian / ...)\n```\n\nThree design choices that matter:\n1. **Single registry, two transports.** HTTP and stdio both call `createMcpServer()` from `mcpServerFactory.ts`. There's no \"stdio has feature X that HTTP doesn't\" — the v2.13.0 drift is closed.\n2. **Sessions are local files.** `~/.vibe-coding-mcp/sessions/.json` — no remote DB, no telemetry, you own the data. Trivial to back up, grep, or sync via Dropbox.\n3. **Publishing is per-platform, not one-size-fits-all.** Notion expects blocks, GitHub Wiki expects sidebar markdown, Obsidian expects frontmatter — `muse_normalize_for_platform` handles each one's quirks so `muse_publish_document` can stay a single tool.\n\n## Configuration\n\n| Env var | Required for | Default | Purpose |\n|---------|--------------|---------|---------|\n| `ANTHROPIC_API_KEY` | `muse_auto_tag`, AI summarization | — | Optional. Enables Claude-API-powered analysis |\n| `NOTION_API_KEY` + `NOTION_DATABASE_ID` | Notion publishing | — | Notion integration |\n| `GITHUB_TOKEN` + `GITHUB_REPO` | GitHub Wiki publishing | — | Wiki push uses git over HTTPS |\n| `CONFLUENCE_BASE_URL` + `CONFLUENCE_USERNAME` + `CONFLUENCE_API_TOKEN` | Confluence | — | Atlassian Cloud |\n| `SLACK_WEBHOOK_URL` | Slack | — | Webhook URL |\n| `DISCORD_WEBHOOK_URL` | Discord | — | Webhook URL |\n| `PORT` | HTTP mode only | `3000` | Streamable HTTP bind port |\n\nAll env vars are optional. The MCP boots with zero env vars set — tools that require an integration return a structured `INTEGRATION_NOT_CONFIGURED` error instead of crashing.\n\n## Known limitations\n\n- **Capture is hook-driven.** The MCP itself doesn't sniff your editor — it relies on Claude Code calling it (via the `PostToolUse` / `Stop` hooks shown in Quickstart). Without hooks, you have to call `muse_session_history` manually.\n- **No shared session store yet.** Sessions are local. Two laptops = two session histories. Cross-device sync is on the roadmap.\n- **AST analysis covers TS / Python / Go.** Rust, Swift, Kotlin, Ruby AST support is not in v2.x.\n- **Publishing is one-way.** We can write to Notion / Wiki / Obsidian etc. but we don't pull updates back into the session store.\n\n## When NOT to use this\n\n- If you want **automatic git commit messages**, use [aicommits](https://github.com/Nutlope/aicommits) or similar — vibe-coding-mcp generates *documents*, not commit messages.\n- If you want **team-wide knowledge base search across everyone's sessions**, this MCP is single-user/local. Build a Notion DB on top and search there, or wait for the shared-store roadmap item.\n- If you want a **manual journaling tool with rich editing**, use Obsidian / Notion directly — this MCP is for the auto-capture-then-publish flow, not for hand-writing notes.\n\n## Comparison\n\n| | vibe-coding-mcp | Manual journaling | `git log` | Notion ADR template |\n|---|---|---|---|---|\n| Captures *why* (decisions), not just *what* | yes | yes (if you do it) | no | yes (if you do it) |\n| Survives skipping a day | yes (hooks) | no | n/a | no |\n| Code-block + AST analysis | yes | no | no | no |\n| Auto-publishes to 6 platforms | yes | no | no | Notion only |\n| Local-first session store | yes (`~/.vibe-coding-mcp/`) | varies | local | cloud |\n| Per-platform Markdown quirks handled | yes | no | n/a | n/a |\n\n## Roadmap\n\n- **Shared session store.** Opt-in cross-device sync (S3 / Git / WebDAV) and team mode where multiple users can search each other's captured sessions.\n- **Auto-summary by topic.** Group sessions by topic across weeks (\"everything I did on the auth refactor in Q2\") instead of by day.\n- **Git commit-message integration.** Surface `muse_summarize_design_decisions` output as a prepare-commit-msg suggestion.\n- **Wider AST language support.** Rust + Swift + Kotlin AST + Mermaid diagrams.\n- **Resource for cross-session search.** `vibe-coding://search?q=...` so the LLM can `@-mention` \"all sessions about caching\" in one go.\n\n## Contributing\n\nPRs welcome. Adding a tool is now a single-file change in `src/tools/.ts` + a one-line `register()` in `src/core/toolRegistry.ts`. CI runs `typecheck + build + test` on Node 20 against every PR.\n\nQuick contributor loop:\n\n```bash\ngit clone https://github.com/MUSE-CODE-SPACE/vibe-coding-mcp\ncd vibe-coding-mcp\nnpm install\nnpm run typecheck && npm test && npm run build\n```\n\n## Security\n\nThis package validates paths, sanitizes filenames, enforces HTTPS-only + allowlisted hosts for webhook URLs, and wraps outbound HTTP with a timeout + exponential-backoff retry — see [`src/core/security.ts`](./src/core/security.ts). The threat model, supported versions, and vulnerability reporting process are in [`SECURITY.md`](./SECURITY.md). CodeQL with the `security-and-quality` query pack runs on every push and weekly.\n\n**Report vulnerabilities privately** via a GitHub Security Advisory: .\n\n## License\n\nMIT — SPDX identifier declared in [`package.json`](./package.json) (a top-level `LICENSE` file will be added in the next release).\n\n## Maintainer\n\n[@yoon-k](https://github.com/MUSE-CODE-SPACE) (MUSE-CODE-SPACE). Issues + support: .\n\n---\n\n\n## 한국어 요약\n\nvibe-coding-mcp는 Claude(또는 다른 MCP 클라이언트)와 함께 **AI 페어 코딩 세션을 자동으로 문서화**해 주는 MCP 서버입니다. Claude Code의 `PostToolUse` + `Stop` 훅에 연결하면 매 세션이 `~/.vibe-coding-mcp/sessions/` 에 자동 저장되고, **15개 도구**가 거기서 **README / DESIGN / API / ARCHITECTURE / CHANGELOG / TUTORIAL** 6종 문서를 생성한 뒤 **Notion / GitHub Wiki / Obsidian / Confluence / Slack / Discord** 6개 플랫폼에 발행합니다.\n\nv2.14.0(2026-05-20)에서 HTTP/stdio 두 진입점이 **단일 레지스트리** 를 공유하도록 통합되었고, `@-mention` 가능한 **MCP Resources 3개**(`sessions/list`, `sessions/{id}`, `config`)와 **Prompts 3개**(`daily-vibe-log`, `document-session`, `refactor-context`)가 추가되었습니다.\n\n설치 + 자동 캡처 훅 설정은 [`docs/AUTO_CAPTURE.md`](./docs/AUTO_CAPTURE.md) 참고. 변경 이력은 [`CHANGELOG.md`](./CHANGELOG.md), 위협 모델은 [`SECURITY.md`](./SECURITY.md).\n",
"bytes": 16419,
"sha": "2e010cd858e3df5bc23ad8ef0abd724c6c0ae4469668ba81a434749ce39f920e",
"repo_slug": "muse-code-space/vibe-coding-mcp",
"fonte": "repo",
"truncated": false,
"api": "https://agentalog.com/api/listings/mcp_io_github_muse_code_space_vibe_coding_403d3b53/readme"
}