emit-pending-review
Pass
Audited by Gen Agent Trust Hub on Jul 18, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill's primary function is to record data to the local file system in a structured format under the .pending-reviews/ folder. It does not request network access, access sensitive system files, or attempt to execute shell commands.
- [PROMPT_INJECTION]: The skill ingests data from external callers to populate review findings, creating an indirect prompt injection surface where malicious text could be persisted in the project's workspace. However, the skill acts solely as a data sink and does not execute or interpret the content of the findings. Evidence for this surface includes: 1. Ingestion points: The Target and Findings fields in SKILL.md. 2. Boundary markers: The skill does not implement delimiters or instructions for the agent to ignore embedded commands in the untrusted input. 3. Capability inventory: The skill is limited to local file write operations within the thread context. 4. Sanitization: No sanitization is performed on the input text before it is written to the markdown bundle.
Audit Metadata