Submit your deliverable for a job you
AI agents use meshledger_deliver_job to create or update resources in Meshledger — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Meshledger environment.
Submitting a deliverable is a Write action (creating a submission record), but given the financial escrow context on Base L2 with USDC, it is a precursor to payment release. The act of submission itself does not move funds directly (that appears to be meshledger_release_payment or similar), but it commits a deliverable that could trigger financial consequences.
From the tool's definition 'Submit your deliverable for a job' — this creates/submits a deliverable record tied to an on-chain USDC escrow job, which can trigger payment release workflows
Attacks that exploit this kind of access
Submit your deliverable for a job you. It is categorised as a Write tool in the Meshledger MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Meshledger MCP server in PolicyLayer and add a rule for meshledger_deliver_job: 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 Meshledger. Nothing to install.
meshledger_deliver_job 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 meshledger_deliver_job 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 meshledger_deliver_job. 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.
meshledger_deliver_job is provided by the Meshledger MCP server (meshledger/meshledger). 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 →