codebase-analysis

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill is designed to automatically invoke a Python script (python3 -m skills.codebase_analysis.analyze) when the user requests codebase understanding. This relies on the execution of shell commands in the agent's environment.
  • [DYNAMIC_EXECUTION]: The analysis workflow involves running Python modules located in the .claude/skills/scripts directory to orchestrate its exploration and synthesis phases.
  • [INDIRECT_PROMPT_INJECTION]: The skill is a tool for ingesting and analyzing external codebases, which creates a surface for indirect prompt injection if the analyzed files contain malicious instructions meant to subvert the agent's behavior.
  • Ingestion points: Any file or directory within the codebase being analyzed (referenced in README.md and SKILL.md).
  • Boundary markers: The provided instructions do not specify any delimiters or safety prompts to ensure the agent ignores instructions found within the target files.
  • Capability inventory: The skill possesses the capability to execute Python scripts and shell commands through its primary orchestration script.
  • Sanitization: No sanitization, filtering, or validation logic for the ingested codebase content is present in the skill's documentation or configuration.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 11:24 AM
Security Audit — agent-trust-hub — codebase-analysis