implement
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill performs shell operations including Git commands (worktrees, commits) and project-specific test or build commands. These are used to verify task completion and manage the workspace.
- [INDIRECT_PROMPT_INJECTION]: The skill processes a variety of untrusted inputs which could contain malicious instructions designed to subvert the agent's behavior.
- Ingestion points: Untrusted repository files, test/build outputs, documentation, and subagent reports (identified in
SKILL.mdandreferences/subagents.md). - Boundary markers: The
Security ModelinSKILL.mdexplicitly instructs the agent to treat these inputs as data, not instructions. - Capability inventory: File system modification, shell command execution (Git and project-defined scripts), and subagent orchestration (identified in
SKILL.md). - Sanitization: Relies on instructional boundaries; there is no mention of automated escaping or filtering of the untrusted content.
Audit Metadata