Skills
skills/ruvnet/claude-flow/agent-matrix-optimizer/Gen Agent Trust Hub

agent-matrix-optimizer

Fail

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONPROMPT_INJECTION
Full Analysis
  • Indirect Prompt Injection (HIGH): The skill is designed to process untrusted external matrix data while possessing the capability to execute code. 1. Ingestion points: matrixData and sparseValues inputs in the analyzeMatrix and solve tools. 2. Boundary markers: None present to delineate data from instructions. 3. Capability inventory: Access to mcp__flow-nexus__sandbox_execute for Python execution. 4. Sanitization: No sanitization or validation of matrix properties before processing is described.
  • Dynamic Execution (MEDIUM): The skill uses mcp__flow-nexus__sandbox_execute to run Python code generated at runtime. While the environment is a sandbox, the dynamic construction of scripts using externally influenced variables like MATRIX_SIZE without validation presents a risk of command injection into the sandbox environment.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 16, 2026, 02:58 AM