agentmako
starsforks
How to install agentmako?
drhalto/agentmako
$git clone https://github.com/drhalto/agentmakoLocal-first MCP server that gives coding agents structured context packets, code/schema facts, and diagnostics - backed by a local SQLite store.
README.md
| 1 | <p align="center"> |
| 2 | <img src="apps/web/public/agentmako.png" alt="agentmako logo" width="180" /> |
| 3 | </p> |
| 4 | |
| 5 | # agentmako |
| 6 | |
| 7 | [](https://www.npmjs.com/package/agentmako) |
| 8 | [](https://github.com/drhalto/agentmako/actions/workflows/smoke.yml) |
| 9 | [](./LICENSE) |
| 10 | [](https://nodejs.org/) |
| 11 | [](https://glama.ai/mcp/servers/drhalto/agentmako) |
| 12 | |
| 13 | agentmako is a local-first codebase intelligence engine for AI coding |
| 14 | tools. |
| 15 | |
| 16 | It gives agents like Codex, Claude Code, Cursor, and local harnesses a |
| 17 | compact Reef-first tool surface for understanding a project before they edit |
| 18 | it. Mako indexes your repo, builds local SQLite-backed facts, tracks |
| 19 | diagnostics and review notes, and answers evidence-backed questions instead of |
| 20 | making the agent orchestrate broad tool chains or rediscover everything with |
| 21 | raw grep. |
| 22 | |
| 23 | Mako is built for the first mile of coding-agent work: |
| 24 | |
| 25 | > What files matter? What routes, symbols, tables, diagnostics, and prior |
| 26 | > findings are relevant? What should the agent read next? |
| 27 | |
| 28 | ## What You Get |
| 29 | |
| 30 | - MCP server for coding agents: `agentmako mcp` |
| 31 | - Local dashboard: `agentmako dashboard` |
| 32 | - Primary project query: `reef_ask` across code, database, findings, |
| 33 | diagnostics, instructions, freshness, and literal checks |
| 34 | - Queryable workflow orientation: `mako_help` |
| 35 | - Deterministic context expansion: `context_packet` |
| 36 | - `_hints` on tool results so agents get result-specific next steps |
| 37 | - Central MCP annotations so clients can distinguish safe reads, live reads, |
| 38 | and local-state mutations |
| 39 | - Compact loop/fallback tools: `reef_status`, `reef_verify`, `reef_impact`, |
| 40 | `live_text_search`, `lint_files`, and `tool_batch` |
| 41 | - [Reef Engine](./docs/reef-engine.md) facts and findings across indexed, |
| 42 | working-tree, and staged state |
| 43 | - Reef convention extraction for auth guards, runtime boundaries, generated |
| 44 | paths, route patterns, and schema usage |
| 45 | - TypeScript, ESLint, Oxlint, Biome, and staged git diagnostic ingestion |
| 46 | - Hot-reloaded `.mako/rules` YAML rule packs, including primitive |
| 47 | cross-file helper-bypass rules via `canonicalHelper` |
| 48 | - Optional Postgres/Supabase schema snapshots and read-only DB inspection |
| 49 | - Local DB review comments for notes on tables, RLS, triggers, |
| 50 | publications, subscriptions, and replication |
| 51 | - Recall, acknowledgements, and agent feedback for repeated review work |
| 52 | |
| 53 | Everything important runs locally. No hosted service is required. |
| 54 | |
| 55 | ## Install |
| 56 | |
| 57 | Requires **Node.js 20 or newer**. |
| 58 | |
| 59 | ```bash |
| 60 | npm install -g agentmako |
| 61 | ``` |
| 62 | |
| 63 | Confirm the CLI is available: |
| 64 | |
| 65 | ```bash |
| 66 | agentmako --version |
| 67 | agentmako doctor |
| 68 | ``` |
| 69 | |
| 70 | You should see green checks for configuration and the local API service. |
| 71 | |
| 72 | > Prefer to build from source (e.g. to contribute)? See |
| 73 | > [Develop From Source](#develop-from-source) at the bottom of this |
| 74 | > file. |
| 75 | |
| 76 | ## Happy Path Setup |
| 77 | |
| 78 | ### 1. Attach your real project |
| 79 | |
| 80 | Go to the project you want Mako to understand: |
| 81 | |
| 82 | ```bash |
| 83 | cd C:/path/to/your/project |
| 84 | ``` |
| 85 | |
| 86 | Attach and index it: |
| 87 | |
| 88 | ```bash |
| 89 | agentmako connect . --no-db |
| 90 | ``` |
| 91 | |
| 92 | Use `--no-db` for the first run. It gets the code intelligence path |
| 93 | working before adding database scope. |
| 94 | |
| 95 | ### 2. Confirm Mako sees the project |
| 96 | |
| 97 | ```bash |
| 98 | agentmako status . |
| 99 | agentmako tool list |
| 100 | ``` |
| 101 | |
| 102 | Run a real Reef query: |
| 103 | |
| 104 | ```bash |
| 105 | agentmako --json tool call . reef_ask "{\"question\":\"where should I inspect auth route state?\"}" |
| 106 | ``` |
| 107 | |
| 108 | If that returns an evidence-backed answer, facts, findings, or next queries, |
| 109 | the core setup is working. |
| 110 | |
| 111 | `reef_ask` plans over code, database, durable findings, diagnostics, and exact |
| 112 | literal evidence. App-flow questions favor file, route, and finding evidence; |
| 113 | RLS/schema questions favor database facts and review comments. To inspect |
| 114 | project rules of thumb directly: |
| 115 | |
| 116 | ```bash |
| 117 | agentmako --json tool call . project_conventions "{}" |
| 118 | ``` |
| 119 | |
| 120 | ### 3. Configure your MCP client |
| 121 | |
| 122 | Add this to your MCP client config: |
| 123 | |
| 124 | ```json |
| 125 | { |
| 126 | "mcpServers": { |
| 127 | "mako-ai": { |
| 128 | "command": "agentmako", |
| 129 | "args": ["mcp"] |
| 130 | } |
| 131 | } |
| 132 | } |
| 133 | ``` |
| 134 | |
| 135 | Restart the MCP client and confirm the `mako-ai` server starts. |
| 136 | |
| 137 | In the agent, default to `reef_ask`. The compact starting surface is: |
| 138 | |
| 139 | - `reef_ask` for project questions across code, database, findings, |
| 140 | diagnostics, freshness, and quoted literal checks |
| 141 | - `reef_status` for maintained issues, changed files, stale diagnostics, and |
| 142 | watcher/schema health |
| 143 | - `reef_verify` for the completion gate over diagnostic freshness and open loops |
| 144 | - `reef_impact` for changed-file blast radius and convention risks |
| 145 | - `mako_help` for an ordered workflow recipe with prefilled arguments |
| 146 | - `live_text_search` for exact current-disk regex/glob inventories |
| 147 | - `lint_files` for bounded diagnostics and `.mako/rules` findings |
| 148 | - `tool |
People who liked this also viewed
6 picksType
- fastapi_mcpExpose your FastAPI endpoints as Model Context Protocol (MCP) tools, with Auth!MCP ServersNov 2025
- lean-lsp-mcpLean Theorem Prover MCPMCP ServersJul 2026
- rulesyncA Utility CLI for AI Coding AgentsMCP ServersJul 2026
- FunASROpen-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.MCP ServersJul 2026
- repomix📦 Repomix is a powerful tool that packs your entire repository into a single, AI-friendly file. Perfect for when you need to feed your codebase to Large Language Models (LLMs) or other AI tools like Claude, ChatGPT, DeepSeek, Perplexity, Gemini, Gemma, Llama, Grok, and more.MCP ServersJul 2026
- mcp-serverMCP server for AI-assisted development of CAP applicationsMCP ServersJul 2026
