merge
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill generates and executes several shell commands for Git operations (checkout, merge, tag, branch deletion) and invokes a Python script (
session_manager.py). The commands use string interpolation for parameters such as{session-id},{winner}, and{agent-id}. If these variables contain shell metacharacters (e.g.,;,&&,|), they could lead to arbitrary command execution. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface where untrusted data (like the winning agent's name or a session ID generated by a previous process) is ingested and used to construct executable shell strings.
- Ingestion points: Variable interpolation in
SKILL.mdfor session IDs and agent names. - Boundary markers: Absent; there are no instructions to validate or escape special characters in these variables.
- Capability inventory: Direct shell access for Git operations and Python subprocess execution.
- Sanitization: Absent; the skill does not include logic to sanitize input before passing it to the shell.
Audit Metadata