write-incident-report
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to process data from potentially untrusted sources, which could be used to influence the agent's output or behavior through embedded instructions.
- Ingestion points: The skill explicitly directs the agent to gather evidence from 'code, logs, tests, deployment records, or user reports' (SKILL.md).
- Boundary markers: There are no specified delimiters or instructions to treat the gathered evidence as passive data, which may lead the agent to follow instructions hidden within that data.
- Capability inventory: The agent has the capability to read files using tools like 'ls', 'sed', and 'rg', and to create new markdown files in the repository's 'docs/incidents/' directory (SKILL.md).
- Sanitization: The workflow lacks steps for validating or sanitizing the content of logs or user reports before they are incorporated into the final incident report.
Audit Metadata