starlight-eng-review
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external technical documentation, source code, and schemas, which creates a potential surface for indirect prompt injection.
- Ingestion points: As described in
SKILL.md, the skill processes inputs such as architecture diagrams, source evidence, state models, and API schemas. - Boundary markers: The prompt instructions do not specify explicit delimiters or "ignore embedded instructions" warnings for the external data being analyzed.
- Capability inventory: The workflow utilizes MCP tools to inspect actual source code, dependency versions, CI configurations, and deployment topologies.
- Sanitization: The skill relies on human oversight, explicitly stating that the review is advisory and requires human approval before any merge action.
Audit Metadata