run-reviewable-subtask-loop
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is a legitimate development tool for managing git workflows. It incorporates extensive security-conscious instructions, prioritizing user consent and repository-defined safety rules for all operations.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from the repository environment, which defines its actions.\n
- Ingestion points: Reads repository instructions, implementation plans, and CI configuration files (
SKILL.md).\n - Boundary markers: Implements a strict 'Enforce consent boundaries' protocol and requires explicit opt-in for the workflow.\n
- Capability inventory: Manages Git branch states (commit, push, merge) and executes local shell-based verification commands.\n
- Sanitization: Requires explicit authorization for remote triggers and verifies branch ownership before cleanup or recovery actions.
Audit Metadata