agent-production-operations
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill implements a feedback loop that processes production traces and user feedback, creating an attack surface for indirect prompt injection.
- Ingestion points:
references/trace-to-eval-feedback.mdsamples production traces, user reports, and incident logs to generate new evaluation cases. - Boundary markers: The skill documentation mandates human review and PII scrubbing of all sampled data, which serves as a mitigation against automated processing of malicious payloads.
- Capability inventory: This skill consists of documentation and schemas; it does not contain scripts that perform automated execution or sensitive system modifications based on the ingested data.
- Sanitization: The feedback pipeline requires scrubbing names, emails, addresses, and secrets before data enters the evaluation suite.
- [COMMAND_EXECUTION]:
AGENTS.mdidentifies the use of internal validation scripts (scripts/validate-skills.rb,scripts/validate-evals.py) during the development and testing phase. These are standard maintenance tools for the agent environment.
Audit Metadata