production-incident-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides a robust set of instructions for analyzing production incidents. It enforces best practices such as service restoration before root cause analysis and requires evidence for all conclusions. It includes evaluation test cases to ensure the agent correctly handles both complete and incomplete data without fabricating information.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze external data provided by the user, which could potentially contain malicious instructions.
- Ingestion points: The prompt instructions in prompts/production-incident-analysis.md describe processing user-supplied alerts, logs, metrics, and deployment records.
- Boundary markers: The instructions explicitly require an 'Input Audit' to verify completeness and credibility, and mandate that unsupported claims be labeled as hypotheses.
- Capability inventory: The skill is strictly analytical and does not use any tools for network communication, file system modification, or command execution.
- Sanitization: The instructions focus on logical auditing and evidence checking rather than technical escaping of the processed text.
Audit Metadata