Загрузка видео...

Не удалось загрузить видео

На главную

THIS MIGHT BE THE #1 OPEN-SOURCE REPO FOR CLAUDE CODE RIGHT NOW. IT GIVES CLAUDE A MEMORY AND SLASHES YOUR TOKEN COST ON EVERY QUESTION The repo is safishamsi/graphify, a free open-source skill that turns any codebase into a knowledge graph Claude Code can read instantly. Instead of grepping...

56,568 просмотров • 5 месяцев назад •via X (Twitter)

Комментарии: 17

Фото профиля Gipp 🦅
Gipp 🦅5 месяцев назад

Cost reduction is what I need

Фото профиля Yarchi
Yarchi5 месяцев назад

you’re welcome

Фото профиля Ridark
Ridark5 месяцев назад

Few channels treat their audience with this much respect

Фото профиля Yarchi
Yarchi5 месяцев назад

🫡

Фото профиля rvaniaaa
rvaniaaa5 месяцев назад

Saving tokens is really important

Фото профиля Yarchi
Yarchi5 месяцев назад

fact

Фото профиля linie
linie5 месяцев назад

that’s some efficiency hack anthropic don’t want us to know about

Фото профиля Yarchi
Yarchi5 месяцев назад

hahah probably

Фото профиля Safi
Safi5 месяцев назад

Bet

Фото профиля rari
rari5 месяцев назад

knowledge graphs make claude feel less forgetful

Фото профиля slash1s
slash1s5 месяцев назад

woe looks so futuristic haha

Фото профиля Nitish Mutha ⚡️
Nitish Mutha ⚡️5 месяцев назад

query a graph not a file tree. this is what context engineering actually looks like. structure your context before claude ever reads it. the bigger the codebase, the bigger the advantage

Фото профиля Roda
Roda5 месяцев назад

Honestly Graphify is the piece that made the whole thing click for me. The graph is derived from your actual structure instead of guessed, and it generalizes way past code. I've been pointing it at an Obsidian vault instead of a repo, and on notes it stops being a code map and becomes memory. A closeout skill writes durable notes after each session, Graphify maps how they connect, and the agent reads the right ones before the next task. Less "navigate my repo," more "remember what we already decided."

Фото профиля SLOT.WIN
SLOT.WIN5 месяцев назад

i just installed it and my token usage already feels like a fair game at my casino every bet counts when the house isn’t burning through chips on every hand

Фото профиля techVector
techVector5 месяцев назад

dude this is actually huge. claude with memory and saving tokens? finally someone made it work. gonna test it out today.

Фото профиля Asher Crowe 🪺
Asher Crowe 🪺5 месяцев назад

the number one repo right now has a shelf life of roughly a tuesday and we all just agree to keep saying it

Фото профиля Insomnia
Insomnia5 месяцев назад

this is what I rly expect

Похожие видео

I can't f*cking believe people still use Claude Code like a chat bot. the ones getting real work out of it gave it a folder, tools and a map of everything it knows. that map is graphify. you point it at a folder and it builds a knowledge graph of everything inside. code, docs, PDFs, images, even video. then your agent searches the graph instead of opening files one by one. 134k stars on GitHub. backed by Y Combinator. and most people still haven't tried it. what makes it different: -> the code part runs fully on your machine. nothing gets uploaded -> no vector database, no embeddings. it's a real graph your agent walks through -> every connection is labeled: read straight from the source, or figured out by graphify. you always know which -> you get 3 files: a clickable map you open in your browser, a report with the key ideas and surprising links, and the full graph your agent queries anytime on a public memory benchmark it found the right answer in its top 10 results about half the time. mem0 managed under 5%. and building the graph cost 0 model credits. here's the full setup around it: > create a folder called AGENT, open Claude Code inside it, and add a CLAUDE.md with one line: "save every decision, person and project to memory/" > run uv tool install graphifyy, then graphify install, then type /graphify . and the whole folder becomes a map Claude can search in seconds > run npm install -g @googleworkspace/cli, then gws auth setup and gws auth login. Claude can now read your Gmail, Calendar and Drive > run claude mcp add playwright npx @playwright/mcp@latest. it gets its own browser for sites, forms and data > run npm i -g jev-guard, then jev-guard key "...", then jev-guard install claude. Jev checks every action before it runs. rm -rf gets blocked, emails with hidden instructions get flagged, about 4 cents per 1,000 checks > type /schedule and say "every morning at 7:30, send me what came in, what you drafted and what needs my yes" one folder. 15 minutes. and Claude finally knows where everything is. all the repos: -> -> -> -> bookmark this and try graphify first. it's the piece everyone skips.

kiosa

31,514 просмотров • 22 дней назад

A DEVELOPER CONNECTED CLAUDE CODE TO OBSIDIAN SO HIS AI AGENT WOULD STOP FORGETTING THE PROJECT EVERY MORNING. Every coding session used to start the same way. Claude would understand the repo, fix the bug, explain the architecture, and then the moment the session ended, all of that context disappeared. Same codebase. Same decisions. Same architecture. Same mistakes repeated again. So he added a memory layer. Instead of treating Claude Code like a smart terminal, he connected it to a local Obsidian vault through MCP. Now Claude can read the repo, open the vault, create notes, link concepts, and write important decisions back into the system. When it studies the codebase, it does not just answer once and forget. It creates notes for the major services, maps how the architecture works, links auth to the database, connects APIs to storage, and records why certain migrations or design choices exist. Obsidian becomes the project graph. Now when he asks why something was built a certain way, Claude does not guess from the current prompt. It reads the decision notes. When he starts a new branch, Claude checks the active context file. When the work is done, it updates what changed, what is blocked, and what the next agent needs to know before touching the repo. That is the real loop: read context, write code, capture decisions, update memory. Most people are still using AI coding tools like disposable chat windows. Ask, patch, close, forget. This setup turns Claude Code into infrastructure. The repo gets a memory layer that survives every session, and multiple AI agents can work from the same project map without stepping on each other. The unlock is not better prompting. The unlock is giving the agent somewhere to remember what it already learned.

DegenCalls

20,124 просмотров • 4 месяцев назад

install these before your next long Claude Code session if your Opus 5.5 bill hurts and you don’t know where to start, copy this: claude-code-router: sits in front of Claude Code and sends each request to the cheapest model that can do it. Haiku does the grep, Opus only gets the hard part. 37k stars rtk: shrinks test, build and git output before the model sees it. one binary, up to 90% less bash output in your context. 83k stars claude-mem: remembers what happened in your past sessions and feeds back only the relevant part, so you stop paying to re-explain your project every morning. 100k stars context-mode: a huge tool output goes into a local database instead of your context window, and Claude pulls out only the lines it needs. 18k stars caveman: makes Claude answer in short telegraphic sentences. code stays identical, and output is the expensive side on Opus 5.5 ($20 per million). 51k stars token-optimizer: a local dashboard that shows where your tokens actually went, and keeps your work alive across compactions. 1.9k stars token-savior: replaces whole-file reads with symbol lookups, so Claude reads the function and not the 2,000 lines around it code-review-graph: maps your repo as a graph and gives Claude only the code connected to the change you’re reviewing jev-ultrafast: a browser agent where Jev picks every click, and a text model wakes up only when something has to be typed. 16,758 stars none of them makes Opus smarter. every one stops it from paying for work it never needed to do the full setup with scripts is in the article. do the measuring step first, so you know what you actually saved

rvaniaaa

25,052 просмотров • 1 месяц назад

STOP BURNING YOUR OPUS 5.5 LIMIT FOR NOTHING these 8 repos cut what Claude reads, writes and re-reads. copy the list before your next long session: headroom: shrinks tool output, logs and files before they reach the model, and keeps the originals so Claude can pull them back when it needs them. 75k stars claude-context: semantic search over your whole codebase, so Claude fetches the relevant snippets and doesn’t read whole files to find them. 12k stars lean-ctx: a local layer that compresses file reads and shell output. a file Claude already read costs about 13 tokens the second time. 3.8k stars openwolf: hooks that keep a project index and session notes, so Claude skips files it already read and doesn’t lose its work at compaction. 2.1k stars claude-token-efficient: one CLAUDE.md that keeps answers short. nothing to install, and output is the $20-per-million side of Opus 5.5. 6.1k stars jcodemunch-mcp: indexes the repo once, then Claude asks for one function and gets just that. it claims 95% less code reading. 2.5k stars ccusage: reads your local Claude Code logs and shows tokens and cost by day and by session. 19k stars Claude-Code-Usage-Monitor: live usage bars and a forecast of when you’ll hit your limit. 8.6k stars none of them makes Opus better. each one just stops it from paying for tokens it didn’t need start with the last two: see where your limit actually goes, then pick two tools from the top. the full setup with scripts is in the article

rvaniaaa

12,381 просмотров • 1 месяц назад

Holy sh*t I spent weeks turning Claude Code into a design studio. no Figma license. no agency. no plugins I had to pay for. here's the stack I used and it's embarrassingly simple: -> graphify: teach Claude your brand dump your guidelines, logos, PDFs and old projects into one folder. graphify turns it into a knowledge graph, so every design Claude makes follows your rules instead of guessing. 134k stars, Y Combinator. uv tool install graphifyy -> graphify install -> /graphify . -> Figma-Context-MCP: Figma straight to code it hands your actual Figma file to the agent. the README says it builds designs in one shot, and does it better than pasting screenshots. -> Penpot: the open-source Figma free, open source, and its MCP server lets AI read and edit your designs, then hand them to code and back. -> React Bits: the "expensive" look, for free the biggest open-source library of animated React components. those smooth hover effects and text animations you see on high-end sites? copy, paste, done. -> HyperFrames: designs that move write HTML, get a finished video. made by HeyGen for AI agents. your landing page can now have its own launch video by tonight. claude plugin marketplace add heygen-com/hyperframes -> claude plugin install hyperframes@hyperframes -> MCP for Blender: 3D without learning Blender connects Claude to Blender. describe a scene, the agent builds and edits it in real 3D. the order that works: brand into graphify -> design in Figma or Penpot -> Claude builds it with React Bits -> HyperFrames makes the promo -> Blender adds the 3D hero shot the uncomfortable part: tools used to be the moat. now everyone has them for free. the only thing that still separates good design from AI slop is taste. and taste is the one thing you can't install. bookmark this. your next client will ask why you're suddenly so fast.

kiosa

73,962 просмотров • 21 дней назад

Why your AI gets dumber the longer you chat (context rot): the one Claude Code file, plus one habit, that fixes it How context rot works: 1. Everything in your chat sits in the AI's working memory, the context window: every message, every file it opened, every answer it gave. The model reads through all of it before every reply. 2. Anthropic, the company behind Claude, says the more tokens you pile into the context window, the worse the model gets at recalling any one thing in it. They call it an attention budget, and every new token spends a little of it. 3. When the window fills up, Claude Code compacts: it replaces the conversation with a summary. If a rule only lived in the chat, it can get lost in that summary. The fix is one file, CLAUDE.md, plus one habit. 4. CLAUDE.md is a plain text file in your project folder. Claude Code reads it at the start of every session, and after compaction it re-reads it from the file. Think of it as a sign bolted above the whiteboard: wipe the board all you want, the sign stays. 5. How to create a CLAUDE.md: open Claude Code in your project folder and type /init. It writes a first draft. Then add what it can't guess: who you are, who the client is, and your rules ("never change a headline without asking", "prices come from column D of the Google Sheet", "dates are day, month, year"). 6. Keep CLAUDE.md short. Anthropic's Claude Code guide says to aim for under 200 lines, because a bloated file gets ignored too. For every line, ask: would it make a mistake without this? If not, cut it. The other half of the fix is how you work: 7. One job, one session. Type /clear when you switch jobs: the board's wiped, and the sign (CLAUDE.md) is still up. 8. If you've corrected the AI twice on the same thing, skip round three in capitals. Clear it, put the rule in CLAUDE.md, and start again. This is context engineering for people who don't code: an AI assistant that remembers your business rules, your clients and your formats, for client work, reports, content, spreadsheets and proposals, whether you're a freelancer, an agency or a small business. #claudecode #aiagents #contextengineering #learnai

Professor Glitch

12,200 просмотров • 1 месяц назад