writing-plans

Pass

Audited by Gen Agent Trust Hub on Jun 26, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection (Category 8).
  • Ingestion points: It processes technical "specs" or "requirements" as the primary input for generating plans (as defined in SKILL.md).
  • Boundary markers: The skill does not utilize specific delimiters or instructions to prevent the agent from obeying malicious instructions embedded within the input specification.
  • Capability inventory: The skill is designed to write plan files to the local filesystem (docs/superpowers/plans/) and generate shell commands including 'pytest', 'git', and 'bash' for subsequent execution by subagents or the user.
  • Sanitization: There is no evidence of validation, escaping, or filtering of the external specification content before it is interpolated into the planning workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 26, 2026, 05:36 PM
Security Audit — agent-trust-hub — writing-plans