concurrent-cli-agents
Fail
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: Instructions provide a command to install the Grok CLI by piping a script from x.ai (a well-known service) directly to bash. This follows established vendor installation practices.
- [COMMAND_EXECUTION]: The
agent-worktree-create.shscript executes shell and git commands to set up isolated worktrees. These actions are transparent and aligned with the skill's stated purpose. - [INDIRECT_PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it manages agents that process untrusted data from git repositories.
- Ingestion points: File
SKILL.mdidentifies that agents process repo content and task descriptions. - Boundary markers: Absent; no explicit delimiters are used to separate instructions from untrusted data.
- Capability inventory: Script
agent-worktree-create.shand instructions inSKILL.mdshow capabilities for file system modification and executing agent CLI tools. - Sanitization: Absent; the skill relies on the end-user's choice of agents and their internal safety mechanisms.
Recommendations
- HIGH: Downloads and executes remote code from: https://x.ai/cli/install.sh - DO NOT USE without thorough review
Audit Metadata