AI agents call source_files to retrieve information from Rr without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
This tool retrieves and enumerates source file information from the debugger's internal state. It performs no modifications, deletions, code execution, or external operations—it is purely informational. The blast radius of misuse is minimal, as an AI agent listing source files cannot compromise system integrity or cause harm beyond information disclosure, which is already presumed in a debugging context.
From the tool's definition Tool name 'source_files' combined with description 'List all source files known to the debugger' indicates a query/list operation that retrieves metadata without modification or side effects.
Attacks that exploit this kind of access
List all source files known to the debugger. It is categorised as a Read tool in the Rr MCP Server, which means it retrieves data without modifying state.
Register the Rr MCP server in PolicyLayer and add a rule for source_files: 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 Rr. Nothing to install.
source_files 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 source_files 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 source_files. 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.
source_files is provided by the Rr MCP server (jnjaeschke/rr-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 →