RemembrallMCP – code dependency graph for agents and memory
Tree-sitter dependency graph saves 5,000-20,000 tokens per agent query vs exploration.
AI coding agent that edits symbols, not strings. AST surgery, full LSP, and a live code graph wired to memory that resurfaces by file, co-change, and semantics.
Four-tier symbol extraction beats reading whole files, but Cursor and Cody already own this space.
Backend developers working with large codebases
Cursor · Continue · Sourcegraph Cody
Tree-sitter dependency graph saves 5,000-20,000 tokens per agent query vs exploration.
74% token reduction + passive agent memory — solves two real pains with zero cloud.
Visual code constellation for AI agents runs locally via single npx command.
Deterministic cross-file impact analysis that catches breaking changes LLMs and linters miss.
Multi-agent task coordination with shared state where Cursor and Claude can actually see each other.
Tree-sitter + SQLite graph reduces agent context 74% while staying entirely local.