explain-then-fix
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from external bug reports and reproduction cases.
- Ingestion points: The agent is instructed to reproduce issues using the "reporter's own case: their data, their environment, their steps" as described in
SKILL.md(Step 2). - Boundary markers: The instructions do not specify any boundary markers or delimiters to isolate untrusted reporter data from the agent's instructions.
- Capability inventory: The skill grants the agent the ability to execute project-specific tools for reproduction and screenshots (
SKILL.md, Step 2) and to modify source code files during the fix phase (SKILL.md, Step 5). - Sanitization: There is no mention of sanitizing, escaping, or validating the external content provided by the reporter before the agent processes it.
Audit Metadata