shrimp-task-manager

Pass

Audited by Gen Agent Trust Hub on May 6, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is composed entirely of instructional Markdown and YAML configuration. No executable scripts (.py, .js, .sh) or binaries are included.
  • [SAFE]: No remote code execution patterns, external downloads, or package installations (npm/pip) were detected.
  • [SAFE]: No sensitive data exposure or exfiltration patterns were found. The skill does not access credentials, private keys, or environment variables.
  • [SAFE]: No obfuscation techniques (Base64, zero-width characters, homoglyphs) or prompt injection attempts were identified.
  • [SAFE]: The instructions include safety guardrails that prevent the agent from fabricating requirements and explicitly state that the tool provides guidance rather than autonomous execution.
Audit Metadata
Risk Level
SAFE
Analyzed
May 6, 2026, 07:20 AM
Security Audit — agent-trust-hub — shrimp-task-manager