Back to browse
Sourceminder.org

Sourceminder.org

by ebcode·Jul 24, 2026·3 points·1 comment

AI Analysis

●●●BangerBig BrainWizardrySolve My Problem

Tree-sitter indexing cuts AI token costs by 15%—actual benchmarks included.

Strengths
  • Tree-sitter parsing means precise symbol lookup, not grep-style string matching.
  • WASM browser demo lets you test queries without installing anything.
  • 15% token reduction is measured and reproducible, not marketing fluff.
Weaknesses
  • Only 7 languages supported; Rust and Perl still marked as provisional.
  • CLI-focused; no IDE integration yet for developers who want in-editor search.
Target Audience

Backend developers, AI engineers building code-aware agents

Similar To

Sourcegraph · grep · ripgrep

Post Description

Hey HN,

The code indexing tools released in December now have added capabilities (Rust, Perl support), better token efficiency, easier install method, and a website! The website has a wasm port of the query tool, qi, so you can try it out in the browser.

Let me know what you think. Thanks!

Similar Projects

Developer Tools●●●Banger

Srclight – Deep code indexing MCP server (FTS5 and Tree-sitter)

Tree-sitter + FTS5 + MCP = tokens saved for AI agents to actually code, not search.

WizardryBig BrainZero to One
srclight
105mo ago
Developer Tools●●Solid

Noodles – Turn any codebase into a diagram with Claude and Tree-sitter

Tree-sitter + agentic flow ~50x faster than raw LLM calls, but codebase visualization is crowded category.

Solve My ProblemShip It
unslop
605mo ago
Developer Tools●●●Banger

Local, privacy-first MCP code intelligence in Rust

AST-aware codebase search and onboarding beats grep, ships as local MCP server.

Big BrainSolve My ProblemWizardry
avirajkhare
104mo ago