{
  "markdown": "# Canvas LMS MCP Server (`canvas-mcp`)\n\n<!-- mcp-name: io.github.admin978/canvas-mcp -->\n\n[![CI](https://github.com/admin978/canvas-mcp/actions/workflows/ci.yml/badge.svg)](https://github.com/admin978/canvas-mcp/actions/workflows/ci.yml)\n[![PyPI](https://img.shields.io/pypi/v/canvas-local-mcp)](https://pypi.org/project/canvas-local-mcp/)\n[![Python](https://img.shields.io/pypi/pyversions/canvas-local-mcp)](https://pypi.org/project/canvas-local-mcp/)\n[![License](https://img.shields.io/github/license/admin978/canvas-mcp)](https://github.com/admin978/canvas-mcp/blob/main/LICENSE)\n\nAsk Claude about your Canvas courses, assignments, deadlines, modules, and grades from one place.\n\n`canvas-mcp` is a local-first MCP server for Canvas LMS users (students, instructors, and MCP builders). It turns Canvas REST API actions into MCP tools that work from Claude Code, Claude Desktop, and other MCP-compatible clients.\n\n> **Status:** alpha. Single-user, no warranty, API surface may still shift. File issues if it breaks.\n\n## Who this is for\n\n- **Students** who want one view across multiple courses\n- **Educators** who want faster access to assignments, modules, and announcements\n- **MCP users** who want Canvas data in local Claude workflows\n\n## What you can ask Claude\n\n- “What assignments are due this week across all my active courses?”\n- “Show upcoming events and planner items for next week.”\n- “List my current grades by course.”\n- “Get the modules (with items) for course `12345`.”\n- “Show announcements for course `12345`.”\n- “Fetch the page `syllabus` from course `12345`.”\n\n## Quick start\n\nPrerequisite: **Python 3.10+**.\n\n### 1) Create a Canvas personal access token\n\nIn Canvas: **Account → Settings → Approved Integrations → + New Access Token**.\nCopy the token shown (Canvas does not show it again later).\n\n### 2) Install\n\nFrom PyPI (recommended):\n\n```bash\npip install canvas-local-mcp\n```\n\nOr from source:\n\n```bash\ngit clone https://github.com/admin978/canvas-mcp.git && cd canvas-mcp\npython3 -m venv .venv && source .venv/bin/activate\npip install -e .\n```\n\n### 3) Configure `~/.canvas.env`\n\n```bash\ncurl -fsSL https://raw.githubusercontent.com/admin978/canvas-mcp/main/.canvas.env.example -o ~/.canvas.env\nchmod 600 ~/.canvas.env\n# edit ~/.canvas.env: set CANVAS_BASE_URL (institution root, no /api/v1)\n# and paste the token into CANVAS_TOKEN\n```\n\nFor token safety guidance (least privilege, file permissions, rotation/revocation, and vulnerability reporting), see [SECURITY.md](SECURITY.md).\n\n### 4) Register in your MCP client\n\nClaude Code:\n\n```bash\nclaude mcp add canvas-local -- canvas-local-mcp\n```\n\nClaude Desktop:\n\n- macOS example path: `~/Library/Application Support/Claude/claude_desktop_config.json`\n- Windows example path: `%APPDATA%\\Claude\\claude_desktop_config.json`\n- Linux example path: `~/.config/Claude/claude_desktop_config.json`\n\n```json\n{\n  \"mcpServers\": {\n    \"canvas-local\": {\n      \"command\": \"canvas-local-mcp\"\n    }\n  }\n}\n```\n\n## Tools exposed\n\n- `list_courses`\n- `list_assignments`\n- `list_modules`\n- `list_announcements`\n- `get_page`\n- `get_file_info`\n- `get_grades`\n- `planner_items`\n- `upcoming_events`\n- `todo`\n\n## Bulk dump\n\n`canvas-local-mcp-dump` downloads course files and related content for offline indexing.\n\n```bash\ncanvas-local-mcp-dump              # all active courses\ncanvas-local-mcp-dump 12345 67890  # specific course IDs\n```\n\nOutput goes to `./canvas-dump/` by default. Override with `CANVAS_DUMP_DIR=/path/to/dir`.\n\n## Local-first, privacy and token flow\n\n- The server runs locally and uses **stdio** transport with your MCP client.\n- Configuration is read from `~/.canvas.env` (`CANVAS_BASE_URL`, `CANVAS_TOKEN`).\n- Requests go from your local server to the official Canvas API endpoints.\n- No external token broker is used in the request path.\n- Security guidance and disclosure policy: [SECURITY.md](SECURITY.md).\n\n## Demo\n\nNo demo GIF is currently committed yet. You can still verify the workflow quickly:\n\n### Try these prompts\n\nAfter registering the MCP server, ask Claude:\n\n- “List my active Canvas courses.”\n- “What assignments are due this week across all my active courses?”\n- “Show upcoming events and planner items for next week.”\n- “List my current grades by course.”\n\n> Placeholder: add a short terminal/GIF walkthrough here in a future PR.\n\n## Development\n\nRequires **Python 3.10+**.\n\n```bash\npip install -e \".[dev]\"\nruff check canvas_local_mcp tests   # lint\npytest                              # tests run against a mocked Canvas API — no token needed\n```\n\nCI runs lint + tests on Python 3.10–3.13 for every push and pull request.\n\n## Publishing a new release\n\nFollow this order to publish a new version (e.g. `0.1.4`) consistently across PyPI and the MCP Registry.\n\n### 1. Bump the version everywhere\n\nUpdate all three files to the new version string in a single PR:\n\n| File | Field |\n|---|---|\n| `pyproject.toml` | `[project].version` |\n| `canvas_local_mcp/__init__.py` | `__version__` |\n| `server.json` | root `version` **and** `packages[0].version` |\n\nThe CI test `test_server_json_version_matches_package` and the workflow validation step will fail if any of these three disagree.\n\n### 2. Merge the PR into `main`\n\nWait for all CI checks to pass before merging.\n\n### 3. Publish `canvas-local-mcp` to PyPI\n\nThe project uses [Trusted Publisher (OIDC)](https://docs.pypi.org/trusted-publishers/) for PyPI uploads.\nTrigger the PyPI publish workflow (or run `python -m build && twine upload dist/*` if you have credentials configured) **before** creating the tag, so the package is available when the MCP Registry fetches it.\n\n### 4. Create and push the tag `vX.Y.Z`\n\n```bash\ngit tag v0.1.4\ngit push origin v0.1.4\n```\n\nThis triggers the `Publish to MCP Registry` workflow, which:\n\n1. Validates that `server.json`, `pyproject.toml`, and the tag all declare the same version.\n2. Publishes `server.json` to the MCP Registry via GitHub OIDC (no secrets needed).\n\n### 5. Verify\n\n- PyPI: `https://pypi.org/project/canvas-local-mcp/`\n- MCP Registry: `https://registry.modelcontextprotocol.io/?q=canvas-mcp`\n- GitHub Actions: `https://github.com/admin978/canvas-mcp/actions`\n\n> **Important**: never reuse or move an existing tag. If you need to re-publish after a mistake,\n> bump to the next patch version (e.g. `0.1.4`) and start from step 1.\n\n## Contributing, roadmap and support\n\n- [Issues / feature requests](https://github.com/admin978/canvas-mcp/issues)\n- Pull requests are welcome for bug fixes and Canvas workflow improvements\n- Roadmap direction currently lives in open issues and upcoming PRs\n\nIf this project helps you manage Canvas with Claude, please try the flow above and consider giving it a ⭐ so other students and educators can find it.\n\n## License\n\nMIT — see `LICENSE`.\n\n---\n\nBuilt by [AGENTE 404 S.L.](https://www.agente404.com) · admin@agente404.com\n",
  "bytes": 6840,
  "sha": "bfe8223e76f16c2c65eee4bd23fc5b6ce0d9b78fedf1191615365389683c7f68",
  "repo_slug": "admin978/canvas-mcp",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_admin978_canvas_mcp_d4a65b35/readme"
}