jcodemunch-mcp: jCodeMunch MCP is a highly token-efficient server that leverages tree-sitter AST parsing to provide precise GitHub source code retrieval. It drastically cuts AI token costs by allowing agents to fetch only the exact code snippets they need, rather than reading entire files.; deja-vu: Deja-vu is a zero-dependency, local-first binary that indexes and makes searchable the historical sessions of various AI coding agents. It acts as a universal memory layer, allowing agents to recall past solutions and context efficiently, even from before its installation.
Significantly reduce AI token costs in retrieval-heavy development workflows, especially for LLM-powered agents.
Reusing solutions from past agent conversations to avoid re-debugging or re-implementing.