AI agents use crow_accept_bot_invite to create or update resources in Crow — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Crow environment.
This is a Write operation because it creates a new bot integration in the Messages system and modifies authorization state. It is not Read (no data retrieval), Execute (no arbitrary code execution), Destructive (reversible by removing the bot), or Financial.
From the tool's definition The tool accepts a bot invite, which modifies the state of the Messages system by adding a bot and triggering authorization.
Attacks that exploit this kind of access
Accept a Crow Messages bot invite. Adds the bot to your Messages so you can chat with it, and tells the bot you accepted so it authorizes you. Paste the bot invite code the owner shared. It is categorised as a Write tool in the Crow MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Crow MCP server in PolicyLayer and add a rule for crow_accept_bot_invite: 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 Crow. Nothing to install.
crow_accept_bot_invite 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 crow_accept_bot_invite 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 crow_accept_bot_invite. 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.
crow_accept_bot_invite is provided by the Crow MCP server (kh0pper/crow). 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 →