AI agents call get_references to retrieve information from Clawmarks without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool performs a data retrieval operation only. It queries and returns reference relationships between clawmarks in the knowledge graph without creating, modifying, deleting, or executing any operations. The action is purely informational with no side effects, fitting the Read category definition of 'retrieves or queries data; no side effects'.
From the tool's definition Tool description states 'Get all clawmarks that reference or are referenced by a clawmark' — a retrieval operation that queries existing bookmark relationships without modifying any data.
Attacks that exploit this kind of access
Get all clawmarks that reference or are referenced by a clawmark. It is categorised as a Read tool in the Clawmarks MCP Server, which means it retrieves data without modifying state.
Register the Clawmarks MCP server in PolicyLayer and add a rule for get_references: 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 Clawmarks. Nothing to install.
get_references is a Read tool with low risk. Read-only tools are generally safe to allow by default.
Yes. Add a rate_limit block to the get_references 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 get_references. 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.
get_references is provided by the Clawmarks MCP server (mrilikecoding/clawmarks). 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 →