AI agents use agloop_create_checkpoint to create or update resources in Agloop — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Agloop environment.
Creating a checkpoint is a reversible write operation that captures and stores the current system state. It does not delete data (not Destructive), execute arbitrary code (not Execute), move money (not Financial), or merely read existing data (not Read).
From the tool's definition Tool name 'agloop_create_checkpoint' and description 'Create a checkpoint of the current state for recovery' indicate this creates and persists a new data record (checkpoint). The 'COORDINATOR ONLY' restriction suggests administrative write access.
Attacks that exploit this kind of access
Create a checkpoint of the current state for recovery. COORDINATOR ONLY. It is categorised as a Write tool in the Agloop MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Agloop MCP server in PolicyLayer and add a rule for agloop_create_checkpoint: 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 Agloop. Nothing to install.
agloop_create_checkpoint 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 agloop_create_checkpoint 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 agloop_create_checkpoint. 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.
agloop_create_checkpoint is provided by the Agloop MCP server (zebbern/agloop-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 →