{
  "markdown": "# Align CLI\n\n[![npm version](https://img.shields.io/npm/v/@aligndottech/cli.svg)](https://www.npmjs.com/package/@aligndottech/cli)\n[![CI](https://github.com/aligndottech/align-cli/actions/workflows/ci.yml/badge.svg)](https://github.com/aligndottech/align-cli/actions/workflows/ci.yml)\n[![License: MIT](https://img.shields.io/badge/License-MIT-yellow.svg)](./LICENSE)\n[![Node](https://img.shields.io/node/v/@aligndottech/cli.svg)](https://nodejs.org)\n\n**Your AI agents know the code. They don't know the company.**\n\nThe decisions behind the code live in commits, tickets, chat and meetings. Months later nobody\ncan tell what still stands, what conflicts, or why. Align pulls them into one graph your agents\ncheck before they build.\n\n```bash\ncurl -fsSL https://align.tech/install.sh | sh\nalign\n```\n\nThat is the whole thing. `align` sets itself up the first time you run it and asks whether you\nwant it local-only or synced; after that it shows your graph and what to do next.\n\nA standalone binary. No Node, no npm, nothing else to install, and local-only mode works\nfully - on-device embeddings included, running on a WASM backend bundled inside the binary.\nLinux, macOS and Windows, x64 and arm64, glibc and musl. Verified against the release's own\nchecksums, and you can read [install.sh](install.sh) before you pipe it anywhere.\n\nPrefer npm, or already have Node? `npm install -g @aligndottech/cli` (Node 22.16+).\nBinaries for every platform are on the [releases page](https://github.com/aligndottech/align-cli/releases/latest).\n\nMIT. No account needed. Beta, pre-1.0.\n\nRun it inside a git repository. `--local` seeds the graph from your commit history, so you have\nsomething to ask about straight away, and your decisions stay in a SQLite file on your machine.\nNothing is sent to Align. [What touches the network](docs/local-mode.md).\n\nWant sync across machines and cross-tool relationship detection? Drop the flag. `align setup`\nlogs you in, connects your tools via read-only OAuth, and wires up your editors.\n\n## How it works\n\n```\n  Your tools                      Align                       Your agents & you\n  ──────────                   ───────────                  ───────────────────\n  Git, GitHub, GitLab  ─┐                                  ┌─ align ask \"why…\"\n  Jira, Confluence      ├─▶  import  ─▶  decision graph ─┐ ├─ MCP server (inline)\n  Linear, Notion        │    (read-only) (what/why/who)  ├─┤  edit hooks (any agent)\n  Slack, Teams, Zoom   ─┘                  + relationships┘ └─ align check (CI)\n```\n\n1. **Import** pulls decisions out of the tools you already use. Read-only, nothing is modified.\n2. Align links them into a **decision graph**: what was decided, why, who decided it, and how\n   decisions relate (supersedes, conflicts with, depends on).\n3. Your agents **query and check against** it, over MCP, an edit hook, CI, or `align ask`.\n\n## Your agent checks before it writes\n\n`align setup` wires Align in four ways, so context fires whether or not the model thinks to ask.\n\n| | What you get |\n|---|---|\n| **MCP server** | Claude Code, Cursor, Claude Desktop and Windsurf query the graph inline |\n| **Edit hooks** | Prior decisions reach the model before it writes. Claude Code, pi, Gemini CLI, OpenCode |\n| **Editor rules** | A managed block in `CLAUDE.md`, `AGENTS.md` and `.cursor/rules/align.md` |\n| **Shared `.mcp.json`** | One committed file wires up the whole team |\n\nThe hook is **non-blocking and fail-open**. It never denies an edit by default, and if Align is\nmissing, slow or unreachable the edit proceeds untouched. It needs no AI provider key.\n\nCursor and Codex CLI can't do the pre-edit hook, and that's a limit of those hosts. They get the\nother three. Full per-host matrix: [docs/agent-hooks.md](docs/agent-hooks.md).\n\n> The first time Claude Code loads a project with a committed hook, it shows a one-time \"approve\n> hooks\" prompt. Accept it to enable automatic alignment.\n\n## Everyday commands\n\n```bash\nalign ask \"how does our auth work\"   # natural language answer with sources\nalign search \"authentication\"        # ranked list, no AI needed\nalign ask src/auth/session.ts        # a file path finds decisions about that file\nalign import git                     # pull commit history, no token\nalign import docs                    # pull ADRs + your CLAUDE.md/AGENTS.md, no token\nalign import jira --token ...        # add more sources\nalign capture <url>                  # capture one decision from a PR, ticket or thread\nalign check                          # check your staged diff against the graph\nalign context sync                   # write decisions to .align/decisions.md\nalign ratify <id>                     # stand behind an agent-made decision, as a human\nalign push <id>                       # send one ratified local decision to the shared graph\n```\n\n`align ask` needs an AI provider to write prose. Without one it returns the matching decisions\nas a ranked list, which needs no key. Bring your own: Anthropic, OpenAI, Gemini, Groq, Mistral,\nGrok, any OpenAI-compatible endpoint, or a local Ollama.\n[Setting one up](docs/configuration.md#ai-provider).\n\n## Docs\n\n| | |\n|---|---|\n| [Importing](docs/importing.md) | Every source, tokens, flags |\n| [Alignment check](docs/check.md) | Modes, exit codes, CI, the GitHub Action |\n| [Cloud or local-only](docs/local-mode.md) | What runs where, what touches the network |\n| [MCP server](docs/mcp.md) | Editor config, the tools your assistant gets |\n| [Configuration](docs/configuration.md) | AI providers, env vars, auth, self-hosting |\n| [Agent hooks](docs/agent-hooks.md) | Per-host capability matrix |\n| [All commands](docs/commands.md) | Full reference |\n\n## Why bother\n\nIn a published benchmark, giving a coding agent recorded product decisions took decision\ncompliance from 46% to 95%\n([Dillon & Varanasi, arXiv:2605.08112](https://arxiv.org/abs/2605.08112) - a small vendor study,\n8 tasks and 41 decision points, and it isn't our data).\n\nWiring context into an agent is the easy part, and this repo is the open-source version of it.\nThe hard part is the record underneath: what your team actually decided, across every tool, kept\ncurrent.\n\nWant a hand setting it up? I do free 30 minute setup calls:\nhttps://calendly.com/tom-align/setup\n\n## License\n\nMIT, see [LICENSE](./LICENSE). The CLI and MCP server are open source, along with the\n[connector SDK](https://github.com/aligndottech/align-connector-sdk). The hosted gateway is a\nseparate commercial service.\n",
  "bytes": 6477,
  "sha": "13322c231f4acb406669aa43ec7b8b94b772c2580f39e69e64232bf6fee7a8ea",
  "repo_slug": "aligndottech/align-cli",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_aligndottech_cli_91cea117/readme"
}