extract_failures
AI agents call extract_failures to retrieve information from Debug Companion MCP without modifying anything — typically the context-gathering step in research, monitoring, and reporting workflows, before the agent takes action elsewhere.
The tool extracts and returns structured information about pytest failures. No description provided, but the name and server purpose indicate data retrieval without side effects. The sibling tool 'run_pytest' executes tests; this tool processes the results. Confidence is high due to clear read-only semantics, slightly reduced due to empty description.
From the tool's definition Tool name 'extract_failures' suggests parsing and retrieving failure information from test output. Server context describes extracting failure locations and displaying code context, which are read-only data retrieval operations.
Attacks that exploit this kind of access
extract_failures. It is categorised as a Read tool in the Debug Companion MCP MCP Server, which means it retrieves data without modifying state.
Register the Debug Companion MCP server in PolicyLayer and add a rule for extract_failures: 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 Debug Companion MCP. Nothing to install.
extract_failures 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 extract_failures 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 extract_failures. 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.
extract_failures is provided by the Debug Companion MCP server (shanirap/mcp-server). 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 →