AI agents use submit_audit to create or update resources in Oathe — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Oathe environment.
This tool creates new audit entries in the oathe.ai security scanning system. While it doesn't modify existing audit data or execute arbitrary code, it does write audit submission records to the platform. The severity is medium because misuse could result in spam submissions or resource exhaustion of the audit service, but it does not irreversibly destroy data, execute code on external systems, or move money.
From the tool's definition Tool description states 'Submit a third-party skill for a behavioral security audit', which creates an audit record in the system. The term 'submit' indicates data creation rather than retrieval or execution.
Attacks that exploit this kind of access
Submit a third-party skill for a behavioral security audit before installing it. It is categorised as a Write tool in the Oathe MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.
Register the Oathe MCP server in PolicyLayer and add a rule for submit_audit: 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 Oathe. Nothing to install.
submit_audit 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 submit_audit 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 submit_audit. 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.
submit_audit is provided by the Oathe MCP server (oathe-ai/oathe-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 →