mcp-prompts

The dev-intelligence-orchestrator is an advanced development assistant designed to bridge the gap between static analysis tools and adaptive learning systems. It serves as a comprehensive orchestration layer for software engineering tasks, automatically detecting project environments and applying learned configurations to optimize tool performance. By leveraging the Model Context Protocol (MCP), it enables seamless communication between development tools like cppcheck, pylint, pytest, and a centralized prompt management server. This skill is intended for software engineers and DevOps professionals who seek to reduce manual configuration overhead and improve the accuracy of automated quality gates within their development life cycle.

• Automatically detects project languages, frameworks, and build systems including C++, Python, Kotlin, Java, PlatformIO, CMake, and Gradle.

• Performs intelligent build error analysis by parsing logs and applying pattern recognition to suggest actionable recommendations.

• Executes static analysis (cppcheck, pylint) with dynamic flag adjustments based on learned historical successes.

• Implements a self-improving learning loop that captures, evaluates, and reuses optimal configurations for recurring development tasks.

• Supports multi-tiered confidence levels for configurations, ensuring that only validated, high-performing settings are prioritized in automated workflows.

• Facilitates seamless integration with mcp-prompts servers via HTTP APIs or standard CLI interactions for flexible storage backends including file systems and AWS services.

• Ensure the mcp-prompts server is accessible in the environment to enable persistent memory of configuration strategies; it degrades gracefully to default behavior if the server is offline.

• Inputs typically include project directories, log files for build analysis, and target source paths for static analysis.

• The system requires common developer utilities like jq for JSON processing and appropriate analysis binaries (pytest, cppcheck, pylint) installed in the environment.

• Practical workflows start with initial runs that capture baseline performance, followed by subsequent automated optimizations based on gathered success metrics.

• Leverage the provided script wrappers for consistent execution; users can extend the orchestrator by adding custom logic to the existing learning domain hooks for specialized project needs.