writing-plans

Pass

Audited by Gen Agent Trust Hub on Mar 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted external data (project specifications or requirements) to generate implementation plans and code snippets.
  • Ingestion points: Reads external spec documents (identified as [SPEC_FILE_PATH] in the reviewer prompt) to define task steps and code logic.
  • Boundary markers: The instructions lack explicit delimiters or instructions for the agent to ignore potentially malicious commands embedded within the input specifications.
  • Capability inventory: The skill has the capability to generate shell commands (e.g., git add/commit, pytest) and write files to the local directory (e.g., docs/superpowers/plans/ and src/ paths), which are subsequently executed or implemented by other tools or agents.
  • Sanitization: There is no evidence of sanitization or validation of the input specification content before it is interpolated into the generated plan document.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 14, 2026, 12:58 AM