The Agent Skills Directory

[SAFE]: No malicious patterns or security risks were detected. The skill follows its defined architectural pattern to perform static analysis of local source code to generate documentation.
[DATA_EXPOSURE]: The skill reads codebase artifacts including configuration and manifest files to understand system architecture. This access is localized to the user-provided target directory and is required for the skill's primary function. There are no patterns suggesting data exfiltration or unauthorized access to sensitive system paths (e.g., ~/.ssh or credentials).
[INDIRECT_PROMPT_INJECTION]: The skill possesses an ingestion surface for untrusted data as it reads and processes files from the target repository. 1. Ingestion points: The Discovery and Tracing agents read READMEs, package manifests, and source code files. 2. Boundary markers: The instructions do not explicitly include delimiters or 'ignore' instructions when interpolating file content into subagent prompts. 3. Capability inventory: The skill's capabilities are limited to reading local files and writing markdown artifacts to a local directory. It does not have network access or shell execution capabilities. 4. Sanitization: No sanitization of ingested code content is performed. 5. Risk Assessment: The impact is limited to the accuracy and content of the generated documentation; the subagents are tasked with descriptive analysis rather than execution, maintaining a safe posture.

linear-walkthrough