i2r-scope-mode
Pass
Audited by Gen Agent Trust Hub on Sep 16, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: No malicious patterns, command execution, or network exfiltration vectors were identified. The skill is purely instructional and utilizes local JSON files for input and output.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from intake files but applies defensive filtering logic to minimize the impact of malicious content.
- Ingestion points: Processes 01-intake.json and 02-context.json from the local run directory.
- Boundary markers: The output is constrained by a JSON schema and requires explicit confirmation flags.
- Capability inventory: No shell access, network operations, or dynamic code execution capabilities are available to this skill.
- Sanitization: Employs a "stack-swap" boundary test and an anti-creep checklist to filter implementation details and gold-plating from the final scope artifact.
Audit Metadata