lmnt
Pass
Audited by Gen Agent Trust Hub on Jun 28, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses npx to fetch and execute the @useclawlink/cli package from the NPM registry, which is the vendor's official distribution channel for their integration tools.
- [COMMAND_EXECUTION]: Shell commands are used to manage LMNT actions, allowing the agent to find, describe, and execute voice synthesis tasks.
- [CREDENTIALS_UNSAFE]: Authentication tokens are managed and stored locally in ~/.clawlink/credentials.json, which is the standard configuration path for the ClawLink environment.
- [PROMPT_INJECTION]: The skill interpolates user-provided text into shell command arguments, specifically in npx @useclawlink/cli actions lmnt "". Ingestion points: User-supplied task descriptions in SKILL.md. Boundary markers: None present in the command template. Capability inventory: Execution of shell commands via the ClawLink CLI. Sanitization: Instructions recommend the agent confirm with the user before any write operations, but no technical sanitization is defined for the command interpolation.
Audit Metadata