srclight/srclight — Deep code indexing MCP server for AI agents.

Description and use

Srclight is a fully local, GPU‑accelerated MCP server that gives AI agents deep code‑indexing capabilities—hybrid FTS5 + embedding search, call graphs, git blame/hotspots, build‑system analysis, and multi‑repo workspaces across 10 languages via tree‑sitter—so assistants can query and act on real source code without any cloud dependencies. Adoption is straightforward: expose its API/SDK/CLI, integrate it as a standard Model Context Protocol server, and pilot it in internal tooling or agent workflows to achieve reliable, context‑aware code assistance. With recent activity, strong ecosystem signals, and high production‑readiness scores, it is a solid OSS candidate for a serious pilot, pending final review of license, security, and maintainer activity.

Problem

srclight/srclight helps connect AI assistants to real tools and data through a standard protocol.

Outcome

High for an OSS candidate: recent activity, adoption, and ecosystem signals are strong enough for a serious pilot.

Workflow

• connect AI agents to tools
• ship Model Context Protocol servers
• standardize integrations

Evidence

Deep code indexing MCP server for AI agents. 25 tools: hybrid FTS5 + embedding search, call graphs, git blame/hotspots, build system analysis. Multi-repo workspaces, GPU-accelerated semantic search, 10 languages via tree-sitter. Fully local, zero cloud dependencies.

Source