compare-architectures

Pass

Audited by Gen Agent Trust Hub on Feb 23, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes a Python script located at '.claude/skills/bmad-commands/scripts/read_file.py' to ingest architectural documentation. This command is used to read and output file content as JSON for the agent to process.
  • [SAFE]: All operations are performed locally within the defined workspace. The skill generates output in the 'docs/' directory and does not attempt to access sensitive system locations, hardcoded secrets, or external URLs.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it ingests and processes untrusted data from external architecture documents and user-provided requirements. 1. Ingestion points: Reads the file path provided in the 'current_architecture' input and parses the 'new_requirements' string. 2. Boundary markers: The instructions do not specify the use of delimiters or protective headers when interpolating external content. 3. Capability inventory: The skill can read local files through a helper script and write new markdown files to the 'docs/' directory. 4. Sanitization: There is no evidence of content filtering or input sanitization for the ingested data before it is analyzed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 23, 2026, 08:44 AM