ashby

Pass

Audited by Gen Agent Trust Hub on Apr 29, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill directs the agent to download and install the @membranehq/cli package from the public NPM registry. This tool is the vendor's official interface for platform operations.
  • [COMMAND_EXECUTION]: Integration tasks such as authentication, action discovery, and data retrieval are performed by executing shell commands through the membrane CLI.
  • [PROMPT_INJECTION]: The skill identifies an indirect prompt injection surface by processing data from Ashby (such as candidate records and notes) that may contain content from external sources.
  • Ingestion points: Command output from membrane action run (SKILL.md).
  • Boundary markers: No delimiters or specific instructions to ignore embedded content are provided.
  • Capability inventory: Subprocess execution of the membrane CLI (SKILL.md).
  • Sanitization: No data sanitization or validation logic is defined for the received Ashby data.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 29, 2026, 11:07 PM
Security Audit — agent-trust-hub — ashby