#codesearch
have you seen the new bug? it's on buganizer. it's literally on issuetracker. dude it's on monorail. find it on gerrit. you can probably find it on googlesource. check it on luci. look for it on codesearch. it's running on clusterfuzz. just sign up for embark. just bugcrowd it.
January 27, 2026 at 10:29 AM
June 9, 2025 at 7:24 AM
If you're REing something and you see a function with a bunch of weird constant values in it, just stick those into google and github codesearch and there's a decent chance you'll know what you're looking at
May 22, 2023 at 7:33 PM
I sure wish I had online codesearch for the source code of all racket packages, with "find references" support
February 17, 2026 at 1:13 PM
If you know a team trying to lower AI coding usage, point them at TopoAccess. It is free, local, model-agnostic, and built to cut context waste before paid API calls. #SaveTheTokens #LLM #DevTools #OpenSource #CodeSearch #Dev #Claude #Codex #AI #TryItOut github.com/mikeanderson42/TopoAccess
April 28, 2026 at 1:58 PM
Antigravity v1.1.6 is here! 🚀 You can now define custom agents right in Markdown with YAML frontmatter. Plus live streaming codesearch & `/copy `! #CLI #DevTools #OpenSource

https://ift.tt/n6ir7HS 💾 https://ift.tt/n6ir7HS
July 25, 2026 at 1:05 AM
AI coding tools are racing toward bigger context. TopoAccess asks the opposite question: how much context can we avoid sending at all? Ask the repo before the model. #SaveTheTokens #LLM #CodeSearch #OpenSource github.com/mikeanderson42/TopoAccess
April 27, 2026 at 10:41 PM
Bigger context windows made repo dumps easier. They did not make them smarter. TopoAccess helps agents find exact repo facts locally before the model spends tokens guessing. #SaveTheTokens #AI #CodeSearch #DevTools github.com/mikeanderson42/TopoAccess
April 27, 2026 at 9:03 PM
Free, local, no GPU needed: TopoAccess helps coding agents avoid spending tokens on basic repo lookup. Find exact files, tests, commands, and validation paths before the LLM call. #SaveTheTokens #OpenSource #DeveloperTools #LLM #AICoding #CodeSearch github.com/mikeanderson42/TopoAccess
April 27, 2026 at 11:04 PM
Want faster AI coding without buying more usage? Start by sending less noise. TopoAccess gives agents local repo context first, so model calls can focus on code instead of file hunting. #SaveTheTokens #AICoding #OpenSource #DevTools #CodeSearch github.com/mikeanderson42/TopoAccess
April 28, 2026 at 12:43 PM
Coding agents should write code, not burn tokens rediscovering your repo. TopoAccess gives them local repo context first: exact lookup, precise files, tests, commands, and validation paths. Model-agnostic. Cache-free. GPU-free. #SaveTheTokens #CodingAgents #DevTools #AI #OpenSource #CodeSearch #LLM
April 27, 2026 at 8:17 PM
google codesearch hat mich während meines studiums oft äh inspiriert. rip
October 23, 2023 at 12:22 PM
Hot with MSNBC for firing Joy, Angela, Tiffany, and other Black folks? Follow Black news and media sources like:

blackcommentator.com
BlackCommentator.com February 27, 2025 - Issue 1030search codesearch code
weekly commentary dedicated to issues of economic justice, social justice and peace affecting African Americans and the African world
blackcommentator.com
February 28, 2025 at 12:25 PM
Days Since Last NPM Supply Chain Attack: 0

MAL-2026-3394: Malicious code in @gaia-codesearch/gaia-api-typescript (npm)
Published: 5/8/2026, 10:05:46 AM
May 8, 2026 at 11:00 AM
Workspace indexing. Open a project and the AI understands your entire codebase. Semantic search, vectors stored locally. Your code never leaves your machine.

#codesearch #infosec #cybersecurity
May 21, 2026 at 3:45 PM
As codebases and AI-generated code scale, text-based search breaks down.

RepoGrep explores semantic-aware code search — a key building block for better reviews, refactors, and AI-assisted engineering.

app.ami.dev/repogrep

#DeveloperTools #CodeSearch
Repogrep
ultra fast codebase search
app.ami.dev
January 10, 2026 at 7:09 PM
Help spread a free way to cut AI coding waste. TopoAccess asks local repos for exact context before Claude/Codex/Cursor spend tokens guessing. Smaller prompts, faster loops. #SaveTheTokens #AICoding #AI #DevTools #CodeSearch #OSS github.com/mikeanderson42/TopoAccess
April 29, 2026 at 4:18 PM
Yup! Learned this after I realised that I'm getting vastly better results from Gemini in the chat and via Cline, and started digging. Apparently not the first to notice:
Gemini 2.5 pro context window seems to be much less than 1M · Issue #8303 · microsoft/vscode-copilot-release
Type: Bug While using Gemini 2.5 pro in both Ask/Agent mode in Copilot Chat, I generally dont use #codebase since I realise the codesearch that copilot performs is quite poor and doesnt seem to inc...
github.com
December 12, 2025 at 2:47 AM
Codesearch is definitely one feature that I miss! I learned a lot from reading how others implemented similar functionalities.

I use GitHub search a lot now and Co-Pilot is pretty helpful.
November 11, 2024 at 4:27 PM
If your agent needs 40k tokens to find the test command, that is not intelligence. That is waste. TopoAccess moves repo discovery local so models can spend tokens reasoning, not rummaging. #SaveTheTokens #AICoding #LLM #DevTools #CodeSearch #OpenSource github.com/mikeanderson42/TopoAccess
April 27, 2026 at 8:47 PM
TopoAccess is free/open-source and built to save tokens where agents waste them most: finding files, tests, commands, artifacts, and validation paths. Ask the repo before the model. #SaveTheTokens #LLM #CodeSearch #OSS #AItools #ClaudeCode github.com/mikeanderson42/TopoAccess
April 27, 2026 at 11:01 PM
I have worked on Google CodeSearch, Kythe, and GitHub’s cross-reference index. It’s not the case that those teams don’t care. Maintaining those services was always a precarious position (“cost center”) within each org despite very positive vibes from users. Scaling (large corpora, index freshness) …
February 9, 2026 at 12:12 AM
Your agent does not need to spend expensive context learning where auth middleware lives. TopoAccess can answer exact repo questions locally, tool-only, before the model gets involved. Send the model the right context, not the whole repo. #SaveTheTokens #AIcoding #LLM #CodeSearch #OpenSource #Agents
April 27, 2026 at 8:18 PM