phantombuster
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONCREDENTIALS_UNSAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill utilizes
npxto download and execute the@useclawlink/clipackage from the NPM registry during runtime setup and usage. - [COMMAND_EXECUTION]: Core functionality relies on the execution of shell commands through the ClawLink CLI to manage PhantomBuster agents and data.
- [CREDENTIALS_UNSAFE]: The skill documentation explicitly identifies that authentication credentials resulting from the login process are stored in the local file system at
~/.clawlink/credentials.json. - [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data retrieved from PhantomBuster (e.g., agent output, lead lists) which could contain malicious instructions.
- Ingestion points: Data fetched via
npx @useclawlink/cli run phantombustercommands as described inSKILL.md. - Boundary markers: None identified; the instructions do not specify delimiters or warnings for the agent when processing external API responses.
- Capability inventory: Shell command execution via
npxand read/write access to the local user directory for credential management. - Sanitization: No evidence of sanitization or validation of the data retrieved from PhantomBuster before it is presented to the agent.
Audit Metadata