Fetching from the wire…
Public story · 2026-07-20 · high
The MIT-licensed tool builds a Tree-sitter map so coding agents stop reading whole repos to answer one question.
Why now: code-review-graph topped GitHub's daily trending list on July 20.
code-review-graph hit 22.5k total stars on GitHub, adding 1,876 in a single day to top daily trending.
The stakes are token bills and agent accuracy on large codebases. Scanning a full repo per question burns tokens and can pull in code that misleads the model.
The project is MIT licensed, built by tirth8205. It keeps a persistent Tree-sitter map of the codebase, so agents pull only the files relevant to a question instead of scanning everything.
The headline number: an 82x median reduction in tokens per question, tested across six repos, ranging from 38x to 528x. Impact F1 averaged 0.714, with rebuilds finishing in under 2 seconds.
The tool ships 30 MCP tools, covering impact radius, architecture overviews, hub-node detection, and cross-repo semantic search. Version 2.3.7 landed July 18.
GitHub's listing doesn't say who ran the six-repo benchmark. The 82x figure is a claim to watch, not a settled result.
Repo-mapping is becoming the default fix for agent context waste, not a niche optimization. Two graph-indexing tools trending in one week isn't a coincidence. It's aimed at Anthropic's own 8.5:1 input-to-output token ratio.
Each link below shares sources, entities, or timing with this story.
This Python project parses repos into a Tree-sitter AST graph of functions, classes and imports linked by call, inheritance and test-coverage edges, stores it in SQLite under .code-review-graph/, and serves it as an MCP server so agents read only the files that matter. Publish...
Anthropic invented a file convention. It's now shipping GA inside a competitor's product. Nobody wrote a spec, nobody held a standards meeting, it just happened. On July 29, GitHub made agent skills and MCP server support generally available in Copilot code review for all Pro,...
This is the install-this-today story. code-review-graph (3.5K GitHub stars) builds a persistent AST-based knowledge graph of your codebase using Tree-sitter, then queries it at review time to compute the minimal context set Claude actually needs. On a 27,732-file Next.js monor...
Three separate Anthropic changes over about two weeks point the same direction, and none of them announced themselves as a strategy. Claude Code 2.1.238 added claude self-hosted-runner --defer-shutdown-max-min, which keeps serving attached sessions on SIGTERM, parks whatever's...
August 6: Sapiom raised $35M for a router that sends each model call to the cheapest capable model. August 7: Databricks published its internal cost playbook. August 8: Toolport hit Product Hunt with a free MCP gateway cutting tool-definition overhead 96%. Three parties, zero...
Y Combinator CEO Garry Tan open-sourced GStack and the repo hit 10,000 GitHub stars in 48 hours. That makes it one of the fastest-growing dev tools of 2026. GStack is a 23-tool MIT-licensed toolkit that turns Claude Code into role-based agents: CEO, Designer, QA, Release Manag...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.