verification-before-completion
Pass
Audited by Gen Agent Trust Hub on Nov 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill requires the agent to ingest and analyze external data to verify task completion, creating an attack surface for instructions embedded in that data.
- Ingestion points: The skill explicitly directs the agent to read "VCS diff" (Version Control System differences) and "代理报告" (Agent reports) in SKILL.md.
- Boundary markers: The instructions lack specific delimiters or guardrails to prevent the agent from executing instructions potentially hidden within the analyzed diffs or reports.
- Capability inventory: The skill logic assumes the agent can execute shell commands for testing, building, and version control management to satisfy the evidence requirement.
- Sanitization: No protocols are defined to validate or sanitize external content before it is processed by the agent.
- [NO_CODE]: The skill consists solely of markdown instructions and metadata, with no accompanying scripts or executable files.
Audit Metadata