usersketch

Pass

Audited by Gen Agent Trust Hub on May 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to use the membrane CLI for authentication, connection management, and executing actions against the UserSketch platform.
  • [EXTERNAL_DOWNLOADS]: Recommends the global installation of the @membranehq/cli Node.js package. This is a vendor-owned resource matching the author's namespace.
  • [PROMPT_INJECTION]: Indirect prompt injection surface identified where the agent processes untrusted data from external tool outputs.
  • Ingestion points: Data returned to the agent from membrane action list and membrane action run commands in SKILL.md.
  • Boundary markers: Absent; there are no instructions to delimit tool output or ignore instructions embedded within the API responses.
  • Capability inventory: The agent has the ability to execute shell commands via the CLI (documented in SKILL.md).
  • Sanitization: Absent; the skill does not provide methods for validating or sanitizing JSON output before it is processed by the agent's context.
Audit Metadata
Risk Level
SAFE
Analyzed
May 2, 2026, 02:49 PM
Security Audit — agent-trust-hub — usersketch