chaos-results-reporter
Pass
Audited by Gen Agent Trust Hub on Aug 12, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill implements a robust 'Hard-reject rule' preventing the fabrication of metrics or reporting on non-existent data, ensuring high data integrity.
- [SAFE]: Data processing is limited to extracting specific fields from local markdown and YAML files (e.g., hypothesis, verdict, timestamps) as defined in the provided field extraction map. There is no evidence of sensitive data exposure or unauthorized file system access outside the designated
results/chaos/directory. - [SAFE]: The skill uses external references (principlesofchaos.org and ISO standards) solely for conceptual definitions and best practices. No external code is downloaded or executed.
- [SAFE]: The output generation follows a structured template for stakeholder reporting, and no prompt injection or obfuscation patterns were identified in the instructions or metadata.
Audit Metadata