AI agents call canvas_get_discussion to retrieve information from Canvas without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs information retrieval only (listing and reading discussions). It matches the Read category definition: 'retrieves or queries data; no side effects (search, list, get, fetch).' The severity is low because retrieving educational discussion content poses minimal risk even if misused by an AI agent — there are no destructive, financial, or code execution implications.
From the tool's definition Tool description explicitly states 'list discussion topics' and 'read the topic' — both read-only operations with no data modification. The Canvas LMS context confirms this retrieves course discussion data without side effects.
Attacks that exploit this kind of access
Without topic_id: list discussion topics in the course. With topic_id: read the topic. It is categorised as a Read tool in the Canvas MCP Server, which means it retrieves data without modifying state.
Register the Canvas MCP server in PolicyLayer and add a rule for canvas_get_discussion: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Canvas. Nothing to install.
canvas_get_discussion is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the canvas_get_discussion rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.
Set action: deny in the PolicyLayer policy for canvas_get_discussion. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.
canvas_get_discussion is provided by the Canvas MCP server (tylergibbs1/canvas-mcp). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.
Every MCP server has a record like this.
Type a name, get the same breakdown: verified identity, auth posture, risk grade, capabilities, recommended policy.
Teams ship this data inside their own products. See what a licence covers →