AI agents use upsert_task to create or update resources in Todos — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Todos environment.
This tool modifies task data non-destructively. It can create new tasks or update existing ones through metadata merging, which are standard Write operations. Severity is medium because while the tool can alter task state at scale depending on filters/fingerprints, these changes are reversible and the scope is limited to task metadata. No irreversible data destruction, code execution, or financial impact.
From the tool's definition Tool description explicitly states it can "Create or update a task" with shallow merge behavior. The name "upsert_task" (upsert = update or insert) and description confirm write-class behavior—data modification is reversible via subsequent updates or…
Attacks that exploit this kind of access
Create or update a task by stable metadata fingerprint. Metadata is shallow-merged on updates. It is categorised as a Write tool in the Todos MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Todos MCP server in PolicyLayer and add a rule for upsert_task: 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 Todos. Nothing to install.
upsert_task is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.
Yes. Add a rate_limit block to the upsert_task 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 upsert_task. 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.
upsert_task is provided by the Todos MCP server (@hasna/todos). 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 →