Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/vishalsachdev/canvas-mcp/agents-mdgit clone --depth 1 https://github.com/vishalsachdev/canvas-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/vishalsachdev/canvas-mcp/agents-md)<a href="https://agentmods.dev/instructions/vishalsachdev/canvas-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/vishalsachdev/canvas-mcp/agents-md.svg" alt="Measured on agentmods" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.04961 | $0.04961 |
| Opus 5 | $0.02481 | $0.02481 |
| Sonnet 5 | $0.00992 | $0.00992 |
| Haiku 4.5 | $0.00496 | $0.00496 |
Grade A, and why
canvas-mcp AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 449 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Canvas MCP - AI Agent Guide
This guide helps AI agents (Claude, Cursor, Zed, Windsurf, and other MCP clients) effectively use the Canvas MCP server.
Quick Start
Canvas MCP is a Model Context Protocol server that bridges AI assistants with Canvas Learning Management System. It provides tools for students to track their academic work and for educators to manage courses, grade assignments, and communicate with students.
Key capability: The server supports both traditional MCP tool calls and a code execution API. Process bulk operations locally without loading every item into the model’s context.
Authentication
All tools require a valid Canvas API token.
Note: The public hosted server (
mcp.illinihunt.org) has been retired — a public MCP endpoint without an access gate would expose the code-execution tool. Use local (self-hosted) mode below. The HTTP/streamable transport remains supported for self-hosting behind your own authentication; for a shared institutional deployment, see deploy/azure/.
Local (Self-Hosted)
Configure credentials in the MCP server's .env file:
CANVAS_API_TOKEN=your_token_here
CANVAS_API_URL=https://your-institution.instructure.com/api/v1
Students and educators use the same server but have access to different tools based on Canvas API permissions.
Tool Profile (Optional)
Reduce tool overhead by setting a role-based profile. Only tools relevant to the selected role are registered:
# In .env:
CANVAS_ROLE=student # ~37 tools (student + shared)
CANVAS_ROLE=educator # 90 tools (educator + shared)
CANVAS_ROLE=all # Default profile; 96 tools by default, 101 with all feature-gated tools enabled
Or via CLI flag: canvas-mcp-server --role student (CLI flag takes precedence over env var).
Tool Categories
Student Tools
Personal academic tracking uses Canvas "self" endpoints. Shared course-content tools still follow the permissions Canvas grants the student's account.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 449 lines · 4,961 tokens per session scan A 9e2a373826a7
canvas-mcp AGENTS.md is an instructions file published in the GitHub repository vishalsachdev/canvas-mcp (209 stars, last pushed 4d ago), licensed MIT. It adds 4,961 tokens to every session, about $0.0248 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
explain-this CLAUDE.md
Instructions for analyticalmonk/explain-this, covering claude.md, what this repo is, the three skills and how they relate, editing these skills and house constraints (the skills fail if these break).
Tutorial_AwesomeModernCPP AGENTS.md
Instructions for Awesome-Embedded-Learning-Studio/Tutorial_AwesomeModernCPP, covering agents.md, 这是什么, 通用 essentials(所有 agent 必读) and 你来做什么?(按场景路由).
open-dictionary AGENTS.md
Instructions for ahpxex/open-dictionary, covering open dictionary rewrite charter, product framing, core workflow, technical framework and 1. raw ingestion layer.
ab900 CLAUDE.md
Instructions for timothywarner-org/ab900, covering claude.md, repository purpose, architecture, commands and 2. non-ascii punctuation -- must return zero.
canvas-lms-mcp AGENTS.md
Instructions for bruchris/canvas-lms-mcp, covering agents.md — canvas lms mcp server, quick start, run with npx (no install needed), or install globally and architecture.
obsidian-university-workflow CLAUDE.md
Instructions for ABO896/obsidian-university-workflow, covering obsidian university workflow, project structure, rules for working on this project, templater api — always read the docs first and config alignment.