agency-performance-benchmarker
Pass
Audited by Gen Agent Trust Hub on Aug 4, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill is designed to ingest and analyze external performance data, including Real User Monitoring (RUM) metrics and API response payloads. This functionality introduces a surface for indirect prompt injection.
- Ingestion points: The agent processes external data streams from RUM tools and API endpoints as part of the benchmarking workflow described in
SKILL.md. - Boundary markers: The instructions lack specific markers or guidance for the agent to differentiate between benchmarking data and potential malicious instructions embedded within that data.
- Capability inventory: The agent uses results to perform bottleneck analysis and recommend optimizations, which could be influenced by injected instructions.
- Sanitization: No sanitization or filtering of external performance metrics or API response content is defined in the skill logic.
Audit Metadata