brida-reflex-agent-progress-gate
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill fetches live versioning and integration metadata from the vendor's resources (specifically
/reflex/llms.txt,/reflex/agent.md, and/reflex/agent.json) to ensure compatibility with current API and SDK availability. - [INDIRECT_PROMPT_INJECTION]: The skill processes agent execution history and step summaries to evaluate progress. This creates a surface for instructions embedded in logs, but the skill includes mitigations by instructing the agent to use deterministic parsing in host code, normalize only bounded state, and rely on synthetic verification fixtures to validate the integration behavior before production use.
Audit Metadata