evo-writing-plans

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFEPROMPT_INJECTIONNO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface when processing external specifications or requirements.
  • Ingestion points: The skill accepts a 'spec' or 'requirements' document as input in SKILL.md.
  • Boundary markers: The instructions lack delimiters or constraints to prevent the agent from following malicious instructions hidden within the input data.
  • Capability inventory: The skill generates implementation plans containing Python code and shell commands (e.g., git, pytest) saved to the 'docs/evo/' directory, which are specifically designed for execution by the 'evo-executing-plans' skill.
  • Sanitization: No validation or sanitization is performed on the input requirements before they are incorporated into the executable plan.
  • [NO_CODE]: The analyzed skill consists exclusively of markdown-based instructions for the AI agent and does not include any executable scripts, binaries, or software dependencies.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 04:31 AM