Skills
Navigate
skills/athola/claude-night-market/muse-delegation/Gen Agent Trust Hub

muse-delegation

Fail

Audited by Gen Agent Trust Hub on Nov 9, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill provides instructions to download and execute an installation script directly from Meta's development domain (https://dev.meta.ai/install.sh). This is a documented installation method for the Muse Code CLI.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process large volumes of repository data (up to 1M tokens), which creates a surface for malicious instructions hidden in the codebase to influence the agent.
  • Ingestion points: Repository files specified via CLI arguments or inlined into the prompt.
  • Boundary markers: Not explicitly defined in the provided instructions.
  • Capability inventory: The skill utilizes the muse exec command for headless code execution and repository-scale analysis.
  • Sanitization: No specific filtering or sanitization of ingested file content is described.
Recommendations
  • HIGH: Downloads and executes remote code from: https://dev.meta.ai/install.sh - DO NOT USE without thorough review
Audit Metadata
Risk Level
HIGH
Analyzed
Nov 9, 2026, 07:25 PM