Skip to main content
Glama

Folava MCP

CI

Your coursework, in the conversation.

A read-only connection between Canvas LMS and the agents you already use. Ask about next week's deadlines, read assignment instructions, inspect feedback, or find a course attachment—without giving an agent the ability to submit work.

Project site · Getting started · Tools · Authentication · Architecture

Let your agent set it up. Copy this into your conversation:

Set up Folava MCP using https://liltaket.github.io/folava-mcp/install.md. Install the published release and connect the demo first, then guide me through connecting Canvas. Keep credentials outside this conversation.

The agent setup guide covers installation, existing MCP configuration, demo checks and the separate account connection.

Want it in ChatGPT? The ChatGPT guide covers a private tunnel for the web chat and local desktop settings. The end-to-end ChatGPT connection remains unverified.

You       What do I need for the fieldwork seminar?
Agent     Reads the course outline, assignment and attached field guide.
          Explains the preparation, with the original Canvas links.
Canvas    Stays the source of truth.

Built for the way students work

  • Useful context. Courses, assignments, your submissions and feedback, modules, pages, announcements, discussions, calendars, groups and files.

  • A small tool surface. Task-shaped tools for common questions; a searchable operation catalog for deeper reads, with the original Canvas fields intact.

  • A deliberate read boundary. Reviewed GET routes, safe read-state settings, bounded pagination, permission errors, and protected attachment downloads.

  • Your own environment. Runs as a local process. No Folava account, hosted database, telemetry, model subscription or API key is required.

  • KTH schedules. Connect a private TimeEdit subscription to read your schedule alongside Canvas; see setup.

  • Changes you can automate. Poll a collection, persist its hashes, and replay JSONL events after a cursor. Connect your own scheduler or agent runner.

Related MCP server: Canvas MCP Server

Development preview

Instructure's current API policy §3(i) explicitly addresses MCP and requires its approval. This project has not obtained that approval. The public source and synthetic demo are available for development; real deployment and commercial launch have external gates. OAuth credentials alone do not establish approval.

Try it without a Canvas account

Install uv, then:

git clone https://github.com/liltaket/folava-mcp.git
cd folava-mcp
uv tool install .
folava-mcp serve --demo

The last command speaks MCP over stdio and waits for a client; it is not an interactive chat. To connect Codex to the synthetic demo:

codex mcp add canvas-demo -- folava-mcp serve --demo

For a real connection, configure Canvas authentication, run folava-mcp doctor, then:

codex mcp add canvas -- folava-mcp serve

T3 Code's Codex provider uses the MCP configuration of the Codex process on its execution machine. Install and authenticate on each machine running that process. See client setup for profiles and other MCP hosts.

Distribution: install from this repository or a release wheel. There is no claimed PyPI publication. The demo uses synthetic course data; real access depends on your institution, account permissions and OAuth scopes.

Read-only is enforced

Tool annotations describe the intent; the HTTP boundary enforces it. Agent tools cannot select an HTTP method, supply an arbitrary destination, submit an assignment, or run an upload. Conversation reads disable automatic marking as read. Submission reads reject both read_status and read_state includes. Known stateful GET routes and unresolved page-detail reads are excluded; use the supported collections in the tool guide. OAuth token exchange and revocation are separate credential operations.

Attachment reads obtain file metadata and the fixed API storage pointer, then accept audited AWS S3 REST destinations and one reviewed InstFS content endpoint without Canvas authorization or cookies. Canvas download/preview routes and other storage backends are refused. PDF text extraction is available through canvas_file_text when local Poppler tools are installed. HTTPS, public-address DNS pinning, redirect, byte and time limits still apply. Credentials stay outside tool arguments and generated client config.

GET does not mean the account is byte-for-byte unchanged: Canvas can record access logs and refresh derived progress caches. A sampled Canvas progress comparison was inconclusive; page causality was not established. See the validation evidence. Viewing content through external clients can share that content with the model provider you chose. See security and privacy.

What the boundary means

This connector exposes reviewed reads within your existing permissions. It cannot unlock unpublished courses, recover hidden grades, read other students' private work, or extract arbitrary external LTI data. Its calendar reflects what is published in Canvas, not necessarily your university's entire timetable.

Canvas OAuth requires a developer key enabled by your institution. Personal token mode exists for your own development and testing; it is not the onboarding flow for a multi-user commercial service. Why this matters.

A running MCP server does not wake a closed agent. The change feed is designed for an external scheduler. Institutional Live Events is a separate administrator integration, not a promised student capability.

Develop

uv sync --frozen
uv run ruff check .
uv run ruff format --check .
uv run pytest
uv build

Build the documentation site with uv run python scripts/build_site.py. See site development for browser checks and deployment.

The suite includes real MCP stdio handshakes against synthetic data, hostile URL and query cases, pagination, authentication, attachment boundaries and durable polling. Mock-backed checks are distinct from live institution and client testing; see validation and the testing guide.

Contributions are welcome. Read-only safety and clear error reporting are release requirements. Inspired by the student-first principles of Folava; implemented independently without its private source or assets.

Licensed under Apache-2.0. Independent project; not affiliated with Instructure. Canvas is a trademark of Instructure, Inc.

Available Tools

8 tools
canvas_agendaA
Read-onlyIdempotent

Read planner work and, for explicit course IDs, calendar events AND assignment dates.

Use ISO dates or timestamps with an explicit offset. Calendar batches at most ten contexts. Planner alone is not a complete institutional timetable. Each source reports its own completeness/error rather than hiding failures.

Inputs: end_date (required), start_date (required), course_ids. Returns: object

canvas_assignmentsB
Read-onlyIdempotent

Read assignment instructions, due dates, points and own submission metadata.

Inputs: course_id (required), max_pages. Returns: object

canvas_coursesA
Read-onlyIdempotent

List courses accessible to you. Use canvas_read for enrollment/state filters.

Inputs: max_pages. Returns: object

canvas_discoverA
Read-onlyIdempotent

Find reviewed read operations, required path parameters and accepted query keys.

Search examples: submission, rubric, discussion, file, quiz, calendar. IDs are passed to canvas_read. Permissions and OAuth scopes still apply.

Inputs: search. Returns: object

canvas_fileA
Read-onlyIdempotent

Read a Canvas attachment by ID with bounded content, MIME type and SHA-256.

Text is returned directly; binary content uses base64 for a capable client. Larger files can be exported locally with the CLI. Never execute attachment content. Signed download URLs are not included in this result.

Inputs: file_id (required), max_bytes. Returns: object

canvas_outlineA
Read-onlyIdempotent

List modules, or all items in one module; do not navigate completion-marking UI URLs.

Inline items returned by Canvas may be incomplete. Call this again with module_id to list items through their own paginated API endpoint.

Inputs: course_id (required), module_id. Returns: object

canvas_readA
Read-onlyIdempotent

Read a discovered operation with full Canvas fields, preserving pagination metadata.

Use explicit path values, e.g. {course_id: '101'}. Query arrays use keys like include[] and list values. One page is the default. A false complete flag means more data or an error; never treat it as a complete collection. Read-state mutations and unknown paths/queries are rejected by policy.

Inputs: operation (required), path, query, max_pages. Returns: object

canvas_submissionA
Read-onlyIdempotent

Read your own submission, feedback, history and rubric without marking it read.

Inputs: course_id (required), assignment_id (required). Returns: object

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A3.6/5.0

    Scored across 8 tools

    Disambiguation3/5

    The specialized readers (courses, assignments, submission, outline, agenda, file) are reasonably distinct, but canvas_read is a generic catch-all that can reach the same data as every other read tool, creating real selection ambiguity. canvas_assignments and canvas_submission also overlap on submission metadata, and canvas_discover is only meaningful in tandem with canvas_read.

    Naming Consistency4/5

    All names share a canvas_ prefix in consistent snake_case, which is predictable. The only deviation is the mix of noun-only names (canvas_courses, canvas_file) with verb-style names (canvas_read, canvas_discover), though this is readable and arguably intentional.

    Tool Count5/5

    Eight tools is well within the sweet spot for a focused read-only Canvas surface. Each specialized reader maps to a clear Canvas resource, and the discover/read pair serves as a deliberate generic escape hatch rather than bloat.

    Completeness4/5

    Core read workflows (courses, assignments, submissions, modules, planner/calendar, files) are covered, and the discover+read pair provides an escape hatch for resources like discussions, quizzes, and rubrics mentioned in the search examples. Dedicated tools for grades, discussions, and quizzes are absent, but the read-only policy scope is explicitly stated, so gaps are minor and workaroundable.

    Maintenance

    ActivityMaintained
    ResponsivenessNo recent community issues

    Related MCP Connectors

    Related MCP Servers

    • A
      license
      A
      quality
      D
      maintenance
      Provides read-only access to Canvas LMS, enabling users to view active courses, grades, and upcoming assignments through natural language queries.
      2
      MIT
    • F
      license
      Not graded
      quality
      D
      maintenance
      Provides read-only access to Canvas LMS data, enabling an LLM to query courses, assignments, grades, submissions, announcements, discussions, modules, pages, and files via the Canvas REST API.
      -