rocketreach

Pass

Audited by Gen Agent Trust Hub on May 3, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the installation of the @membranehq/cli package from the official NPM registry, which is a standard procedure for this vendor's integrations.
  • [COMMAND_EXECUTION]: Provides several shell commands for the membrane CLI, including login, connect, and action run. These commands facilitate authentication and interaction with the RocketReach API through the Membrane platform.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it ingests and processes data from external RocketReach records which could contain instructions intended to influence the agent.
  • Ingestion points: Output from the membrane action run and membrane action list commands (e.g., contact info, company data).
  • Boundary markers: None are specified in the instructions to distinguish between trusted instructions and untrusted data.
  • Capability inventory: The agent can execute further actions (membrane action run), create new actions (membrane action create), and modify state within the connected RocketReach account.
  • Sanitization: There is no mention of sanitizing or validating the output from the external API before the agent processes it.
Audit Metadata
Risk Level
SAFE
Analyzed
May 3, 2026, 11:22 PM
Security Audit — agent-trust-hub — rocketreach