The Agent Skills Directory

[SAFE]: The skill is designed for API discovery and auditing, which is a standard development and security analysis task.
[DATA_EXPOSURE]: The skill reads project configuration files (such as package.json, requirements.txt, and go.mod) and traverses source code to extract endpoint definitions, authentication requirements, and database dependencies.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection because it ingests untrusted data from the codebase it analyzes.
Ingestion points: The skill reads all files within the codebase across Phases 1, 2, and 4 to identify routes and handler internals.
Boundary markers: There are no explicit delimiters or instructions to ignore embedded malicious prompts within the processed files.
Capability inventory: The agent has the capability to read any file in the codebase and write to local documentation and telemetry files.
Sanitization: The skill does not perform sanitization or validation of the code content before interpolating it into its internal analysis or output files.
[COMMAND_EXECUTION]: The skill utilizes targeted search patterns for specific frameworks (like Express, FastAPI, and Spring Boot) to identify API routes, which constitutes safe and purpose-built code analysis.
[DATA_EXFILTRATION]: No evidence of network exfiltration was detected. The skill writes its findings to docs/api-surface-map.md and appends telemetry data to a local file in ~/.claude/projects/, remaining entirely on the local file system.

api-surface