writing-plans
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data in the form of 'specs' or 'requirements' and decomposes them into executable tasks. If a provided specification contains malicious instructions (e.g., instructions to include a data exfiltration step in the 'minimal implementation'), the skill may incorporate these into the final plan, which is then processed by execution sub-skills.
Audit Metadata