fetch_attachment
AI agents call fetch_attachment to retrieve information from Outline MCP Server without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool name and server purpose indicate this retrieves/downloads an attachment from Outline documents. No modification, deletion, or execution occurs. The empty description lowers confidence slightly, but the semantic meaning of 'fetch' strongly indicates a read-only retrieval operation. Blast radius if misused is low—it would only expose whatever data the attachment contains.
From the tool's definition Tool name 'fetch_attachment' and server context describe a document management system focused on 'document search, reading, and management'. Fetching an attachment is a retrieval operation with no side effects.
Attacks that exploit this kind of access
fetch_attachment. It is categorised as a Read tool in the Outline MCP Server MCP Server, which means it retrieves data without modifying state.
Register the Outline MCP Server MCP server in PolicyLayer and add a rule for fetch_attachment: 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 Outline MCP Server. Nothing to install.
fetch_attachment 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 fetch_attachment 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 fetch_attachment. 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.
fetch_attachment is provided by the Outline MCP Server MCP server (mcp-outline). 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 →