effect-command-executor
Pass
Audited by Gen Agent Trust Hub on May 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent on how to use
ChildProcess.makeand theChildProcessSpawnerservice to execute shell commands and system binaries such asgit,bun, andnode. This is the primary intended function of the skill. - [PROMPT_INJECTION]: The documentation demonstrates the construction of commands using template literal interpolation (e.g.,
ChildProcess.makebun publish ${name}`). This identifies a surface for Indirect Prompt Injection (specifically argument injection) if the agent interpolates untrusted data from external sources into these command templates without proper sanitization. - Ingestion points: Variables interpolated into
ChildProcess.maketemplate strings inSKILL.md. - Boundary markers: None explicitly shown for variable interpolation.
- Capability inventory: Full subprocess execution via
ChildProcessSpawner.spawn,spawner.string, andspawner.linesas described inSKILL.md. - Sanitization: The skill provides technical usage examples and does not explicitly include input validation or sanitization logic for the interpolated command arguments.
Audit Metadata