interactive-work
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted user input, such as bug reports, error logs, and task requirements, which are then used to inform the agent's code investigation and implementation strategies. This creates a vulnerability where malicious instructions embedded in a bug report could influence the agent to perform unauthorized code changes or data exposure.
- Ingestion points: User-provided descriptions of bugs and tasks, including status messages and stack traces (SKILL.md).
- Boundary markers: Absent; the protocol does not utilize delimiters or specific instructions to isolate user input from the agent's core directive to 'Investigate and execute'.
- Capability inventory: The skill is intended to read repository files, find root causes, and apply code modifications or refactors (SKILL.md).
- Sanitization: Absent; there is no guidance for the agent to sanitize, escape, or validate the content of user-provided data before acting upon it.
Audit Metadata