Read a task attachment reported by get_task_by_id. Images are returned as MCP image content when possible.
AI agents call read_task_attachment to retrieve information from Codex Omnifocus without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves attachment data from an already-existing task without altering state. It performs no write, execute, destructive, or financial operations. Reading task attachments is a safe, side-effect-free operation that merely accesses stored information.
From the tool's definition Tool name is 'read_task_attachment' and description states it 'Read[s] a task attachment' and 'Images are returned as MCP image content'.
Attacks that exploit this kind of access
Read a task attachment reported by get_task_by_id. Images are returned as MCP image content when possible. It is categorised as a Read tool in the Codex Omnifocus MCP Server, which means it retrieves data without modifying state.
Register the Codex Omnifocus MCP server in PolicyLayer and add a rule for read_task_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 Codex Omnifocus. Nothing to install.
read_task_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 read_task_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 read_task_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.
read_task_attachment is provided by the Codex Omnifocus MCP server (phd-peter/codex-omnifocus-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 →