omakase-off

Pass

Audited by Gen Agent Trust Hub on May 14, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes shell commands to manage isolated development environments, specifically utilizing git worktree and git branch for creation and deletion of variant-specific contexts as described in Phase 3 and Phase 5 of the detailed workflow.- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it ingests untrusted user requirements and interpolates them into implementation plans and tests executed by subagents.
  • Ingestion points: User-provided build, create, or implement requests processed at the entry gate (SKILL.md).
  • Boundary markers: The skill does not define explicit delimiters or instructions to ignore embedded commands when passing user requirements to the writing-plans dependency.
  • Capability inventory: The skill performs file system operations (writing plans), git repository manipulation (worktrees), and dispatches subagents with execution permissions via parallel-agents (detailed-workflow.md).
  • Sanitization: There is no documented validation or sanitization of user-provided feature descriptions before they are used to generate executable content.
Audit Metadata
Risk Level
SAFE
Analyzed
May 14, 2026, 01:33 AM
Security Audit — agent-trust-hub — omakase-off