The Agent Skills Directory

SAFE (SAFE): No malicious patterns or security vulnerabilities were detected in the provided files.\n
The extract_concepts.py script utilizes the standard ast library for Python analysis, which is a secure method for parsing code without execution.\n
There is no evidence of command execution, network activity, or credential harvesting.\n
The documentation and scripts are consistent with the stated purpose of building ontological models.\n- Indirect Prompt Injection (LOW): As the skill is designed to ingest and process external source code (untrusted data), it possesses an indirect prompt injection surface. However, this is inherent to the functionality of a documentation tool and is categorized as low risk.\n
Ingestion points: The extract_concepts.py script reads source code files from paths provided as command-line arguments.\n
Boundary markers: The script outputs structured JSON-LD or Mermaid diagrams but does not include explicit LLM defensive instructions in its output stream.\n
Capability inventory: The skill allows for file system reading and metadata extraction from local directories.\n
Sanitization: The script extracts class and function names directly from source code and does not sanitize them before including them in the output, which is expected for its use case.

grey-haven-ontological-documentation