AI agents invoke install_local_extension to trigger actions in Todos. What it does depends on the arguments the agent supplies, and its effects often reach beyond the immediate call — builds kicked off, notifications sent, workflows started.
Installing or updating extensions executes system-level operations that modify the runtime environment. This goes beyond a simple write (data modification) — it deploys executable code/workflows that can alter system behavior. An AI agent misusing this could install malicious or unintended extensions with potentially broad impact on the system's behavior.
From the tool's definition "Install or update a local workflow extension from a manifest, directory, or offline bundle"
Attacks that exploit this kind of access
Install or update a local workflow extension from a manifest, directory, or offline bundle. It is categorised as a Execute tool in the Todos MCP Server, which means it can trigger actions or run processes. Use rate limits and argument validation.
Register the Todos MCP server in PolicyLayer and add a rule for install_local_extension: 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.
install_local_extension is a Execute tool with high risk. Execute tools should be rate-limited and have argument validation enabled.
Yes. Add a rate_limit block to the install_local_extension 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 install_local_extension. 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.
install_local_extension 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 →