writing-plans
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill transforms design context into code and tasks, which could be exploited via malicious inputs. 1. Ingestion points: External design context. 2. Boundary markers: None. 3. Capability inventory: Generation of code and shell commands (git, pytest). 4. Sanitization: No sanitization of ingested content.
Audit Metadata