NYC

Dark Matter Analyzer

Fail

Audited by Gen Agent Trust Hub on Feb 15, 2026

Risk Level: HIGHPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [Prompt Injection] (HIGH): The skill possesses a significant Indirect Prompt Injection surface (Category 8). It is designed to process untrusted data (repository contents) and has the capability to modify files and create artifacts. A malicious file within an analyzed repository could contain instructions that redirect the agent's logic to delete data or inject backdoors.\n
  • Ingestion points: Full repository access including code, docs, and metadata.\n
  • Boundary markers: None identified.\n
  • Capability inventory: modifies_files, creates_artifacts.\n
  • Sanitization: No evidence of sanitization or instruction filtering for analyzed content.\n- [Command Execution] (MEDIUM): The documentation promotes the use of dark-matter-analyzer-mcp, an unverified third-party tool. Executing the suggested CLI commands (dark-matter scan) could lead to the execution of untrusted code if the agent attempts to install or run this external dependency.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 15, 2026, 10:54 PM