sync-findings
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from GitHub PR comments and local review artifacts, creating a surface for indirect prompt injection attacks.\n
- Ingestion points: The agent reads untrusted data from local findings records and external GitHub PR content (comments, thread state) as described in the SKILL.md workflow steps 2 and 4.\n
- Boundary markers: The instructions do not define specific delimiters or instructions to prevent the agent from obeying commands embedded within the synced data.\n
- Capability inventory: The skill allows the agent to update local documentation (
path/findings.md) and interact with remote GitHub PR threads (replying to and resolving comments) as described in SKILL.md workflow steps 3 and 6.\n - Sanitization: There is no evidence of content sanitization or validation before the data is written to the findings record or used to determine GitHub thread actions.
Audit Metadata