writing-issues
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from discussions, verbal dumps, and external reviews to generate structured issue tracker items.\n
- Ingestion points: The skill instructions in
SKILL.mdspecify turning "a discussion, review finding, or plan into tickets" and using "a rough verbal dump" as source material.\n - Boundary markers: While the output templates use bold headers (e.g.,
**Problem / Why:**) as structural delimiters, there are no explicit instructions to ignore potentially malicious commands embedded in the source data.\n - Capability inventory: The skill focus is restricted to text generation and refinement; it does not request tool access, shell execution, or network capabilities in its frontmatter or body.\n
- Sanitization: No specific input validation, escaping, or filtering mechanisms are defined to handle malicious content within the provided source materials.
Audit Metadata