skills/ruvnet/ruview/Pair Programming/Gen Agent Trust Hub

Pair Programming

Pass

Audited by Gen Agent Trust Hub on Mar 4, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized data access behaviors were detected. The skill primarily functions as a documentation guide and command reference for a development utility.\n- [EXTERNAL_DOWNLOADS]: The skill documents the installation of the 'claude-flow' package from the NPM registry as a prerequisite for its pair programming features.\n- [COMMAND_EXECUTION]: The skill facilitates the execution of various development-related shell commands through the 'claude-flow' CLI, including test suite execution, git operations, and code quality analysis.\n- [PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface as it analyzes external project data. 1. Ingestion points: Reads and explains local source code, git history, and project structure (SKILL.md). 2. Boundary markers: Employs truth-score verification thresholds and configurable safety agents. 3. Capability inventory: Capable of implementing code, executing tests, and committing changes via CLI integrations. 4. Sanitization: Includes built-in security scanning, real-time code review, and performance analysis metrics to monitor AI-generated outputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 4, 2026, 07:39 PM