analyze-scaling-regime
Pass
Audited by Gen Agent Trust Hub on Sep 24, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is purely informational and instructional. It defines a conceptual procedure for analyzing scaling regimes without incorporating executable scripts, external packages, or commands involving network or file system operations.
- [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing untrusted data.
- Ingestion points: Data enters the agent's context via the
outcome_seriesandobservation_contextparameters specified inSKILL.md. - Boundary markers: The instructions do not specify the use of delimiters or 'ignore' warnings for the data being processed.
- Capability inventory: No risky capabilities (such as network access, file system writes, or subprocess execution) were identified within the skill definition.
- Sanitization: No input validation, filtering, or sanitization steps are defined.
Audit Metadata