alphasense

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of the @membranehq/cli package from the NPM registry. This is a vendor-owned resource used to manage authentication and interface with the Membrane platform.
  • [COMMAND_EXECUTION]: The skill provides a set of instructions to execute membrane CLI commands. These commands are used for user authentication, discovering AlphaSense actions, and performing API requests via a proxy service.
  • [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection as it retrieves and processes data from the external AlphaSense API which could contain malicious instructions.
  • Ingestion points: Responses from AlphaSense API calls executed via membrane action list and membrane request (documented in SKILL.md).
  • Boundary markers: The instructions do not define specific delimiters or instructions to treat external data as untrusted.
  • Capability inventory: The skill can execute various AlphaSense actions and arbitrary HTTP requests through the membrane CLI (documented in SKILL.md).
  • Sanitization: No sanitization or validation of external API content is documented within the skill's workflow.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 08:54 AM