itsol-execution-policy
Pass
Audited by Gen Agent Trust Hub on Sep 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and apply execution policies from external repository files and user-provided instructions, which could be manipulated to influence agent behavior.
- Ingestion points: The policy resolution logic explicitly incorporates
repo-defaultconfigurations and.itsol.mdfiles located within the project repository. - Boundary markers: The skill instructions mandate specific precedence rules for policy resolution and require observable evidence for completion (
done_whencriteria), which serves to limit unintentional behavior but does not fully sanitize the input surface. - Capability inventory: While primarily a governance skill, it authorizes and controls high-level agent capabilities including subagent delegation, resource allocation (CPU/parallelism), and transitions between critical stages such as planning and implementation.
- Sanitization: There is no explicit mechanism defined for sanitizing or validating the schema of the ingested configuration files before they are interpreted as binding agent instructions.
Audit Metadata