The Agent Skills Directory

[SAFE]: The skill is designed for code explanation and documentation purposes. It utilizes the Python ast (Abstract Syntax Trees) module for structural analysis, which allows the agent to inspect code without executing it.
[SAFE]: No malicious patterns, obfuscation, or data exfiltration attempts were detected. The instructions include explicit security boundaries, such as advising the agent to explain code statically unless execution is specifically authorized in an isolated environment.
[SAFE]: External references are limited to well-known educational platforms like Real Python, which provide documentation and tutorials.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it processes user-provided code.
Ingestion points: User-provided code snippets provided via the skill's arguments.
Boundary markers: The skill includes specific 'Implementation limits' that warn against treating output as a substitute for environment validation and prohibit rendering untrusted labels as active markup.
Capability inventory: Capabilities are limited to static analysis (ast, re) and markdown generation; no network or file-writing tools are invoked in the scripts.
Sanitization: The skill relies on Python's ast.parse for analysis, which does not trigger code execution.

code-documentation-code-explain