Context Engine
Compresses noisy logs, RAG results, and code context into high-signal input for AI agents.
Compresses noisy logs, retrieval fragments, and code context into structured, high-signal input for AI coding assistants and agent workflows. Operates in three modes: root-cause extraction from logs, question-aware evidence reranking for RAG pipelines, and minimum-viable-context extraction for code repair.
Source
Repository: https://github.com/melonelish/context-engine
Maintain Context Engine?
Let people know it's listed here — add the badge (live metrics, light/dark aware) or a plain link to your README or docs.
[Context Engine on getagentictools](https://getagentictools.com/mcp/melonelish-context-engine?ref=badge) Related mcp
Coze Workflows Bridges Coze workflow automation with MCP clients for listing, executing, and managing Coze workflows through conversational inter
Trace Framework-aware code intelligence MCP server with 44+ tools for code tracing, analysis, and debugging.
Code to Tree Converts source code into Abstract Syntax Tree representations using tree-sitter parsers for multiple programming languages includ
Cocos Creator Embedded MCP server for Cocos Creator Editor: 89 tools for AI agents to inspect and operate running Cocos projects.
EvalView Regression testing for AI agents with golden baselines, CI/CD integration, and multi-framework support.
Roblox Studio Runtime Controls and debugs live Roblox Studio sessions from local assistant workflows.