tmux-cli
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates data ingestion from other tmux panes which could contain unverified instructions.
- Ingestion points: Captured output from tmux panes using the 'tmux-cli capture' and 'wait_idle' commands described in SKILL.md.
- Boundary markers: No delimiters or specific safety instructions are provided to the agent for handling captured data.
- Capability inventory: The skill allows for the execution of shell commands in other panes via 'tmux-cli execute'.
- Sanitization: The skill description does not specify any sanitization or validation logic for the captured output.
- [COMMAND_EXECUTION]: The skill provides tools to run shell commands in active tmux panes.
- Pattern: 'tmux-cli execute' as documented in SKILL.md.
- Scope: Specifically intended for automation tasks such as builds and testing.
- [EXTERNAL_DOWNLOADS]: The skill instructions include the installation of a vendor-owned external dependency.
- Dependency: 'claude-code-tools' to be installed by the user via 'uv tool install'.
- Context: This utility is necessary for the skill to interact with the tmux environment.
Audit Metadata