acquire-codebase-knowledge

Pass

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a bundled Python script (scripts/scan.py) to gather codebase metadata. This script uses subprocess.run with argument lists to invoke git commands (e.g., git log, git rev-parse), preventing shell injection. \n- [PROMPT_INJECTION]: The skill processes content from the target codebase, which serves as a surface for indirect prompt injection. \n
  • Ingestion points: Source files, configuration, and scan results containing comments and metadata are read by the agent. \n
  • Boundary markers: The instructions lack delimiters or specific instructions to ignore embedded commands in analyzed content. \n
  • Capability inventory: The agent can create files in docs/codebase/ and execute the local analysis script. \n
  • Sanitization: Data from the codebase is processed without filtering or validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 14, 2026, 12:39 AM