road2qnt

[shiron @godofwaaar]

this MIT quant paper is f*cking insane

it tested 100 strategies per run across 10,000 Monte Carlo trials

every strategy had zero real alpha

but after selecting the best backtest:

+38.84% in-sample

then on fresh data:

+0.14%

and 49.16% of the “winners” actually went negative out-of-sample

that is the trap with AI trading

the more strategies you generate, the easier it becomes to manufacture a genius-looking backtest by accident

so the real workflow is:

generate → backtest → select → unseen data → robustness gates → paper trade

the edge is not producing more strategies

it is killing the ones that only look profitable

[Annatar.md @AnnatarXBT]

agent memory, in 13 pages

the fix that took one setup from 26,000 tokens per query down to 1,800

every morning your agent rereads the same 12 files, lands on the same conclusions and bills you twice for it

the 5 layers that end that:
> working memory - what it sees right now, gone once the window fills
> episodic memory - a timestamped log, so the deploy that died at 3am gets explained only once
> semantic memory - facts in a knowledge graph, like "this user prefers TypeScript"
> procedural memory - three approaches tried, the one that worked saved as a skill
> forgetting - what to drop before contradictions pile up

the numbers:
> Mem0 - 1,800 tokens per query instead of 26,000, with 91% lower latency
> Snowflake - one ontology layer, 20% more accuracy, 39% fewer tool calls

forgetting is the layer almost nobody builds, and it decides whether the other four stay worth trusting

the memory system for Opus 5.5 that survives across sessions, step by step, is below 👇

[Quant Science @quantscience_]

Financial Machine Learning (159 page PDF)

Download it here:

[Infallible @infalliblexbt]

This is f*cking insane. Stanford just dropped a 2-hour course on building LLMs from scratch

Don't waste 2 years trying to learn how Claude & ChatGPT are built

• 00:00 - LLM tokenization
• 25:44 - how LLMs decode your prompts
• 35:40 - the LLM training pipeline
• 1:16:47 - LLM architecture from scratch

Anthropic pays engineers $750,000/year for exactly this kind of LLM knowledge

Bookmark it and find 2 hours today, no matter what

Then read the article below ↓

[Chonsy @0xChonsy]

this is f***ing wild

SpaceXAI engineer Lauren Tan (@poteto) shared her method for running agents with GitHub repo skills:

give your agent this to identify bugs before rewriting the code

Prompt:

{
"inputs": {
"repo": "[REPO]",
"task": "[TASK]",
"success": "[OBSERVABLE_SUCCESS]"
},
"mission": [
"Use the smallest justified change.",
"Define observable success before editing.",
"Demonstrate the result in the running system."
],
"investigate": {
"entry_point": "Start with /poteto-mode when available.",
"skills": {
"/how": "Trace current behavior.",
"/why": "Recover design rationale and supporting evidence.",
"/teach": "Connect mechanics and intent.",
"/recall": "Restore relevant prior context."
},
"requirements": [
"Reproduce the problem and preserve the baseline.",
"Separate confirmed evidence from inference.",
"Use only relevant skills supported by the host."
]
},
"design": [
"Use /architect for caller usage, types and module boundaries.",
"Use /arena to compare solutions to the same task.",
"Use /swarm to divide independent work.",
"Define ownership and shared interfaces before parallel edits.",
"Use /interrogate for critical review when useful."
],
"build_verifier": {
"condition": "The repository lacks a working verification path.",
"skill": "/create-verification-skill",
"launch": "Exact startup command and readiness signal.",
"doctor": "Usable instance, dependencies and required access.",
"drive": "Real UI, CLI or API interaction.",
"evidence": "Observed outputs, logs, traces and artifacts.",
"cleanup": "Only processes and data created by this run.",
"feature_map": [
"entry point",
"actions",
"observable result"
],
"trial": "Execute the instructions on at least one real feature and retain evidence."
},
"match_proof": {
"ui": "Complete the changed user journey.",
"cli": "Run the actual command and record its result.",
"storage": "Read back the written value.",
"parser_or_migration": "Process a saved input fixture.",
"performance": "Compare equivalent before-and-after profiles.",
"measurement": "Use /benchmark-checklist to inspect cache state, build parity and measured work."
},
"execute": [
"Implement, run, inspect and correct.",
"Rerun the original failing case after the patch.",
"Check relevant regressions.",
"Check functionality, performance and code quality separately when relevant.",
"For export retries, inspect the produced file and confirm repeated execution does not duplicate rows.",
"Support success claims with observed evidence.",
"Report a concrete blocker when a required check cannot run."
],
"prevent_recurrence": {
"reflect": "Use /reflect to extract reusable lessons.",
"correct": "Use /correct to enforce recurring corrections.",
"priority": [
"Architecture or better data structures.",
"Types, lint or CI.",
"Behavioral regression tests.",
"Written rules for what remains."
],
"validation": "Demonstrate that the original failure is prevented or caught.",
"mechanics": "Move stable mechanical work into reusable scripts."
},
"handoff": {
"return": [
"tested_revision",
"diff",
"commands",
"results",
"evidence_paths",
"unrun_checks",
"remaining_risks",
"blockers"
],
"required_check_status": [
"PASS",
"FAIL",
"NOT_RUN"
],
"reporting": "Attach a result and evidence to each required check. Explain missing checks.",
"review": "Prepare the change under the repository's review and merge rules."
},
"scale": [
"Expand autonomy after repeatable verification.",
"Keep the goal, completion condition and permission boundaries explicit.",
"Define isolation and retain a decision log.",
"Stop with a concrete explanation when blocked.",
"/loop is a Cursor capability.",
"Measure human review time and regressions."
],
"start": [
"Inspect the repository.",
"Restate the task and observable success.",
"Reproduce the baseline.",
"Begin the first bounded step."
]
}SpaceXAI engineer Lauren Tan shared her method for running agents with reusable repo skills:

give your agent this to identify bugs before changing code.

Prompt:

{
"inputs": {
"repo": "[REPO]",
"task": "[TASK]",
"success": "[OBSERVABLE_SUCCESS]"
},
"mission": [
"Use the smallest justified change.",
"Define observable success before editing.",
"Demonstrate the result in the running system."
],
"investigate": {
"entry_point": "Start with /poteto-mode when available.",
"skills": {
"/how": "Trace current behavior.",
"/why": "Recover design rationale and supporting evidence.",
"/teach": "Connect mechanics and intent.",
"/recall": "Restore relevant prior context."
},
"requirements": [
"Reproduce the problem and preserve the baseline.",
"Separate confirmed evidence from inference.",
"Use only relevant skills supported by the host."
]
},
"design": [
"Use /architect for caller usage, types and module boundaries.",
"Use /arena to compare solutions to the same task.",
"Use /swarm to divide independent work.",
"Define ownership and shared interfaces before parallel edits.",
"Use /interrogate for critical review when useful."
],
"build_verifier": {
"condition": "The repository lacks a working verification path.",
"skill": "/create-verification-skill",
"launch": "Exact startup command and readiness signal.",
"doctor": "Usable instance, dependencies and required access.",
"drive": "Real UI, CLI or API interaction.",
"evidence": "Observed outputs, logs, traces and artifacts.",
"cleanup": "Only processes and data created by this run.",
"feature_map": [
"entry point",
"actions",
"observable result"
],
"trial": "Execute the instructions on at least one real feature and retain evidence."
},
"match_proof": {
"ui": "Complete the changed user journey.",
"cli": "Run the actual command and record its result.",
"storage": "Read back the written value.",
"parser_or_migration": "Process a saved input fixture.",
"performance": "Compare equivalent before-and-after profiles.",
"measurement": "Use /benchmark-checklist to inspect cache state, build parity and measured work."
},
"execute": [
"Implement, run, inspect and correct.",
"Rerun the original failing case after the patch.",
"Check relevant regressions.",
"Check functionality, performance and code quality separately when relevant.",
"For export retries, inspect the produced file and confirm repeated execution does not duplicate rows.",
"Support success claims with observed evidence.",
"Report a concrete blocker when a required check cannot run."
],
"prevent_recurrence": {
"reflect": "Use /reflect to extract reusable lessons.",
"correct": "Use /correct to enforce recurring corrections.",
"priority": [
"Architecture or better data structures.",
"Types, lint or CI.",
"Behavioral regression tests.",
"Written rules for what remains."
],
"validation": "Demonstrate that the original failure is prevented or caught.",
"mechanics": "Move stable mechanical work into reusable scripts."
},
"handoff": {
"return": [
"tested_revision",
"diff",
"commands",
"results",
"evidence_paths",
"unrun_checks",
"remaining_risks",
"blockers"
],
"required_check_status": [
"PASS",
"FAIL",
"NOT_RUN"
],
"reporting": "Attach a result and evidence to each required check. Explain missing checks.",
"review": "Prepare the change under the repository's review and merge rules."
},
"scale": [
"Expand autonomy after repeatable verification.",
"Keep the goal, completion condition and permission boundaries explicit.",
"Define isolation and retain a decision log.",
"Stop with a concrete explanation when blocked.",
"/loop is a Cursor capability.",
"Measure human review time and regressions."
],
"start": [
"Inspect the repository.",
"Restate the task and observable success.",
"Reproduce the baseline.",
"Begin the first bounded step."
]
}

[Quant Science @quantscience_]

Quant Bible:

a 51-page guide written by an MIT Sloan Business Club member, covering probability, statistics, market making, and real interview questions from Jane Street, Citadel, Two Sigma and more

all for free

Bookmark this (before they take it down):

[adon @F5Adon]

天哪，Karpathy 在 GitHub 上免费放了一个完整的 AI 学位课程

7 个仓库，306,741 星标，而且按正确顺序学习，它们会带你从反向传播到能通宵运行研究的智能体 。

每个仓库都是前一个的下一层。仓库地址👇

[Сarm1ne @carm1nee]

these 2 github repos are the most BROKEN combo in trading right now

Citadel pays quant researchers $336K to $642K a year to forecast markets and stress-test trades. These 2 repos put both steps inside Claude Code for free.

They're called Kronos and Day Trader. Kronos was built by researchers at Tsinghua University, accepted at AAAI 2026, and it already has 45K+ stars. Day Trader is 11 Claude skills built by Jake Schincariol, and it already has 4K+ stars.

First, get how it works:

1. Kronos gives you 1,000 futures instead of one guess. It reads candlesticks the way ChatGPT reads words, it learned from 12 billion candles across 45 exchanges, and it can map out the next 48 hours of any chart.

2. Day Trader turns that fan into a trade. One skill sets your stop and target before you enter, one sizes the position to your max loss, one blocks anything over your limits, and its Time Machine replays the setup 10,000 times before you risk a dollar.

3. Together they answer the question that matters: how many of those futures hit my target before my stop? On the real BTC candles in the video, 725 of 1,000 futures end down, so the plan is a paper short with a 1% stop and a 2% target. 203 hit the target first, 302 hit the stop first, 495 touch neither. Nothing gets placed until you type yes.

Here's how you can run it for free yourself:

1. Open Claude Code in an empty folder.

2. Install Day Trader. Paste: "https://t.co/UqFeQtR8CM install skill". No signup, no API key.

3. Install Kronos. Paste: "Set up https://t.co/UyxiaNhP0J and load Kronos-small from Hugging Face."

4. Run /trader-scanner and /trader-news to pick a ticker.

5. Ask Claude Code: "Sample 1,000 Kronos futures for this ticker from its last 128 candles, 48 candles ahead. Keep every path."

6. Check it before you trust it: "Re-run Kronos on each of the last 5 days and show me how often the real price landed inside its p10-p90 range."

7. Run /trader-exit to set your stop and target, then ask: "Race every Kronos path to my stop and target. How many hit each one first?"

8. Run /trader-timemachine for a second opinion: 10,000 replays from the stock's own candles, plus every past time the setup fired. If it disagrees with Kronos, skip the trade.

9. Run /trader-budget and /trader-risk, then /trader-broker. It shows you the ticket and sends it to your paper account only after you type yes.

i won't say anything more:

[Divyanshi @Divyyanshishrma]

If I had to land a $200K AI engineer job in 90 days, I would not get a degree.

I would master these 10 GitHub repos.

1. awesome-llm-apps
The production AI playbook. RAG, agents, multimodal apps, all in working code. 106K+ stars.
Repo → https://t.co/50OD473w70…

2. LangChain
The foundational framework. Used in production by Klarna, Replit, Elastic, and most AI startups in 2026.
Repo → https://t.co/e8juC9sbHg…

3. LangGraph
The orchestration layer powering production agents. The skill on every senior AI engineer job description.
Repo → https://t.co/e8juC9sbHg…

4. CrewAI
Multi-agent coordination. The framework most Fortune 500 teams reach for first.
Repo → https://t.co/Ixp7pBR3sD…

5. Ollama
Run any open-source LLM on your own machine. The fastest way to learn how models actually work.
Repo → https://t.co/x4D63YcRCg

6. awesome-mcp-servers
MCP is the standard every major AI lab adopted in 2026. Knowing it puts you ahead of 99% of engineers.
Repo → https://t.co/2sxxKGwmT8…

7. Qdrant
The vector database used for production RAG at scale. Embeddings and semantic search are non-negotiable for AI roles.
Repo → https://t.co/OfhwKdk0zI

8. AI-Agents-for-Beginners
Microsoft's free 12-lesson course on building agents. Real code, real exercises, real prep.
Repo → https://t.co/J2ijcIpCCn…

9. system-design-primer
Production AI is system design. The repo FAANG engineers use to prep for interviews.
Repo → https://t.co/g52zkFf5MB…

10. awesome-claude-code
The playbook for the tool now used inside FAANG, OpenAI, Anthropic, and most YC startups.
Repo → https://t.co/rprwu6qQsp…

Here's the wildest part:

A $200K AI engineer in 2026 isn't paid for a degree.

They are paid for what these 10 repos teach.

The market doesn't care where you learned it. It only cares if you can ship.

90 days. 10 repos. One portfolio that proves you can do the work.

That's it. That's the whole game.

Save this before you forget.

100% free. 100% open source.

[Sammy @sammy_capital]

These REA prompts for Claude, Grok and GPT are insane...

I made one for Opus 5.5, one for Grok 4.7 and one for GPT-6.1 Sol

Each gives the model a clear role, the right REA tools, a loop for checking every claim against evidence and rules for what to hand the next model

They work on almost any target: APKs, Electron apps, .exe files, firmware or websites you're allowed to analyze

Bookmark and send them to your models

Then read the article about turning three AI models into one reverse-engineering lab

[Leandro von Werra @lvwerra]

Another DeepSeek drop: dsh-superagent

An agent framework that works through a repo's GitHub issues and PRs with many agents in parallel, designed to take as little human attention as possible while delivering as much value as possible.

https://t.co/XR73mIvA0V

[Devansh @thenowhereway]

Vibe coders, thank me later

https://t.co/7YwmCmGgq8
https://t.co/28dcoCIKKM
https://t.co/a1iooDf6Fv
https://t.co/AS4LJUQZHw
https://t.co/MhlqsBf1MP

Your app stops looking AI-generated!

[Divyanshi @Divyyanshishrma]

A GRAD STUDENT WITH 32 FOLLOWERS JUST HIT 56,400 STARS.

His name is Yuxiang Lin. GitHub handle: Lum1104.

He's at Georgia Tech. He researches multimodal emotion recognition. He plays badminton in his spare time. His personal site is a .dev subdomain. His pinned bio still says he's "currently working on LLM agents."

A few months ago he had 655 stars on a side project called Understand-Anything — a plugin that turns any codebase into an interactive knowledge graph.

Then everything broke open.

The project got picked up by Egonex (Infinite Universe, Inc.). It went from a personal repo to an org-backed release. It got translated into Chinese, Traditional Chinese, Japanese, Korean, Russian, and Spanish. It hit Trendshift.

56,400 stars later, it's becoming the default code-understanding plugin for nearly every AI coding tool that exists — Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode, OpenClaw, Antigravity, Pi Agent, Vibe CLI, Hermes, Cline, KIMI CLI, and Trae.

Here's what the plugin does:

→ Tree-sitter parses your codebase for the structure
→ LLM agents add the meaning
→ Builds a knowledge graph of every file, function, and class
→ Opens an interactive dashboard you can pan, zoom, and search
→ Generates guided tours so you learn a 200K-line codebase in the right order
→ Tells you what your changes break before you commit
→ Auto-updates on every commit via a post-commit hook

568 commits. 7 releases. MIT license. v2.7.3 shipped May 19, 2026.

One honest note: the LLM passes cost real tokens on large repos. The structural graph is free and reproducible. The semantic layer is where your model bill shows up.

The README still credits him by name. The license still has his name on it. © Yuxiang Lin and Infinite Universe, Inc.

A researcher who studies how machines understand human emotion built the tool that finally teaches machines to understand code.

This is what open source was supposed to feel like.

100% Open Source.

Repo in the first comment.

[Musolsol.𝟎𝐱𝐔 @MMMusol]

大学教授可能没想到，
GitHub 上已经有人把私人家教开源了...

这个项目叫 DeepTutor，由香港大学数据智能实验室开发，目前在 GitHub 上已经获得超过 4.8 万颗 Star！

它不是普通的 AI 聊天机器人，而是一套完整的 AI 个性化学习系统。

你可以把教材、论文、PDF 扔进去，它就能围绕你的资料建立知识库，讲解知识点、解答问题，甚至自动生成练习题，检查你到底学会了没有。

更夸张的是，它支持深度研究、交互式可视化、沉浸式阅读，还能根据学习进度规划掌握路径。

如果你学高数，它可以辅助拆解公式；

如果你学外语，它可以陪你练习；

如果你研究论文，它还能结合资料帮你梳理知识体系。

它甚至支持长期记忆、接入 QQ 和微信等聊天渠道，让 AI 老师不再局限于一个网页。

项目采用 Apache 2.0 开源协议，支持本地部署，不过调用部分 AI 模型可能产生费用。

未来的教育，
也许是每个人都有一个真正了解自己的 AI 老师：
https://t.co/PMhHd3IlZW

[Sammy @sammy_capital]

REA FROM ZERO TO REVERSE-ENGINEERING AGENT

One practical path built only from REA's official docs: guides, real investigations, evidence contracts, clean-room rules and verification checks

START HERE:

1. Your First REA Investigation https://t.co/VMq1vzHMKt
2. Guided MCP Prompts https://t.co/3qqMBrfhPB
3. Installation & Agent Setup https://t.co/xPcibAOnaw

WATCH & FOLLOW ALONG:

1. REA Explained https://t.co/mNig06NZZr
2. How To Actually Reverse Engineer Anything: REA + Claude Code https://t.co/7rnsZtUPJf
3. How to Use REA with AI for Free https://t.co/WN8xm6g2wz
4. Reverse Engineer Anything: Copy an App for Free https://t.co/mOJG1696J3
5. REA on ShareX https://t.co/EHnuqv9R2q
6. What Is REA? Hidden Logic Test https://t.co/fUEWiKlGe0
7. Frida for Android Reverse Engineering https://t.co/xOPHb5akx7
8. Ghidra Tutorial for Beginners https://t.co/nYl41r238P
9. I Tested the Viral REA Skill https://t.co/HQelD3loRX 10. REA Video Collection https://t.co/v4rLy56OYf

READ THE EVIDENCE:

1. MCP Runtime Contracts https://t.co/77IFlTo1XX
2. Evidence Workflows https://t.co/phAF22PYfj
3. Recorded Guide Evidence https://t.co/s52PLxi1ZH

TRACE ONE FUNCTION:

1. Notes Export https://t.co/VMq1vzHMKt
2. Notion Clipboard https://t.co/X6mmOrpyY6
3. DX-Ball Reconstruction https://t.co/MLPT8XZ1G6
4. Native Investigation Guide https://t.co/0EJAHiMauX

CLEAN-ROOM SPEC:

1. Clean-Room Separation https://t.co/kZyKXzZozL

VERIFY THE REBUILD:

1. verify_reconstruction https://t.co/3qqMBrfhPB
2. Reconstruction Readiness https://t.co/coTcnKCcWv
3. Aegis APK Case https://t.co/SqChuIEHRX
4. DX-Ball Verification https://t.co/MLPT8XZ1G6

BEST ORDER TO LEARN:

install REA → Notes Export → read the evidence → Notion → DX-Ball → clean-room spec → Aegis verification

[RazeDen @razeden0]

this shouldn't be legal... i just merged 6 ais into one brain

opus 5.5 + REA + Jev + Dots + Grok Bot + Obsidian

6 agents. 1 shared memory. i gave them $10 and one job: make money before i wake up

the only instruction: find something people already pay for, make it better, sell it. no questions

here's what happened while i slept:

23:40 - laptop closed
00:12 - grok bot reads 3,800 posts on X
00:14 - 1,900 couriers hate the same thing
00:58 - REA takes apart 14 paid apps that try to fix it
01:05 - cheapest is $19/mo. every single one has the same 2 bugs
01:40 - obsidian pulls 1,240 notes and 214 skills from my old projects
02:30 - opus 5.5 plans 9 screens and 1 database. asks me nothing
04:10 - opus 5.5 writes the app, grok bot ships it
04:52 - jev blocks a $49 domain. budget is $10. it picks a $3 one
05:45 - dots cuts a 30 s demo and posts 3 threads
06:41 - 212 signups
07:15 - 31 of them pay $9/mo
07:20 - obsidian writes down every step, every cost, every mistake

7:40 - i wake up

31 x $9 = $279
spent: $4.70 tokens + $3 domain = $7.70
36x back in 7h 40m

the real secret: no agent here can do it alone. grok can't think. jev can't write. opus can't post. linked together they never stop

the big mistake: 6 ais in 6 tabs. 6 memories, 0 results. one shared brain fixed it

you don't need to hire 6 people. you need 6 agents and 1 night ↓

[Hamza Khalid @humzaakhalid]

If you started using Claude this year, install these before anything else.

Here are 32 free skills that skip the months of trial.

They come from 9 GitHub repos with 252K stars.

NOTES & RESEARCH

obsidian-markdown → Notes Obsidian can read.
obsidian-bases → Notes into tables.
json-canvas → Mind maps on a canvas.
defuddle → Clean text from any page.
notebooklm → Query your notebooks.
huggingface-papers → Find AI papers.

WRITING

humanizer → Strip the AI tells.
copywriting → Landing pages and ads.
copy-editing → Edit line by line.
social → Posts for LinkedIn and X.
emails → Write email sequences.
content-strategy → Plan what’s next.
writing-guidelines → Make docs clear.

MARKETING

seo-audit → Find why you don’t rank.
ai-seo → Get cited by AI.
launch → Plan a product launch.
pricing → Set your prices.
lead-magnets → Freebies that grow lists.
marketing-psychology → Why people buy.

AUTOMATE

n8n-workflow-patterns → Proven n8n flows.
n8n-node-configuration → Set up nodes.
n8n-expression-syntax → Fix expressions.
n8n-validation-expert → Read n8n errors.
n8n-code-javascript → Code in n8n nodes.

BUILD & SHIP

react-best-practices → Fast React code.
web-design-guidelines → Catch UI mistakes.
deploy-to-vercel → Go live with a link.
modern-python → Python with uv and ruff.
huggingface-gradio → Quick Python demos.
huggingface-spaces → Host your AI app.
huggingface-local-models → Run models locally.
context-fundamentals → How context works.

Don’t install all 32 tonight.

Pick the folder closest to your daily work and add one skill from it.

Give it a week of real tasks. If it saves you time, keep it and add the next.

Honestly, 3 skills you use daily beat 30 sitting in a folder.

For more cheat sheets like this, subscribe → https://t.co/Hp3qoiIww0

Repost ♻️ to help a Claude beginner in your network.
https://t.co/ebMbmGokWP

[Sammy @sammy_capital]

THE COMPLETE AI REVERSE-ENGINEERING TOOLKIT: 150+ FREE RESOURCES

Agents, free courses, native binaries, .NET, iOS, Electron, browsers, firmware, WASM and protocols, and most builders have no idea it's all free

01 - START WITH REA

Installation and agent setup
https://t.co/xPcibAOnaw
Your first REA investigation
https://t.co/VMq1vzHMKt
REA guides
https://t.co/8k2KaopWvN
Guided MCP prompts
https://t.co/3qqMBrfhPB
MCP runtime contracts
https://t.co/77IFlTo1XX

Also useful: Evidence Workflows · Native Investigation Guide · JavaScript Application Graph · Electron Observation Guide

Install: npx rea-agents@latest setup

02 - REAL REA INVESTIGATIONS

Notes Export - button → IPC → CSV
https://t.co/VMq1vzHMKt
Notion Clipboard - page API → preload → main process
https://t.co/X6mmOrpyY6
DX-Ball - Windows binary → verified C function
https://t.co/MLPT8XZ1G6
Aegis - APK → TOTP implementation
https://t.co/SqChuIEHRX

Also useful: CTF Flag Checker · REA Gems · Chrome Dinosaur investigation · Notes Evidence Record

03 - AI REVERSE-ENGINEERING AGENTS

IDA Pro MCP
https://t.co/wq7tl6tjdq
GhidrAssistMCP
https://t.co/ZNod0siO4z
GhidraMCP
https://t.co/HHS8Kwyhys
Official radare2 MCP
https://t.co/GyhPKQ822n

Also useful: Revenant · Awesome LLM Reverse Engineering · AgentRE-Bench · SRE-Bench · Reverse Engineering Agent Skill

04 - NATIVE BINARIES

Ghidra
https://t.co/P6Abt6uAoH
radare2
https://t.co/UlLiewb4V8
Rizin
https://t.co/Zq9kghSi4N
Cutter
https://t.co/1u6oNXDSwf

Also useful: Keystone · GNU Binutils · GDB · LLDB · pwntools · ROPgadget · Detect It Easy · capa · YARA

05 - .NET APPLICATIONS

ILSpy
https://t.co/bj56i2svw0
dnSpyEx
https://t.co/Is2BcZbDRh
dnlib
https://t.co/EuUHFO131j
de4dot
https://t.co/jEfLZodDAH

Also useful: AsmResolver · Harmony · Mono.Cecil · NativeAOT reversing helpers

06 - iOS

Frida iOS Guide
https://t.co/liJ8vIfaJt
OWASP iOS Testing Guide
https://t.co/rFcSpYih9Z
DVIA-v2
https://t.co/lN3xqrEbZn
libimobiledevice
https://t.co/QpYS12OLfQ

Also useful: ios-deploy · class-dump · ipsw · Objection iOS

07 - ELECTRON AND JAVASCRIPT

Electron ASAR
https://t.co/hfrJuU56DP
Electron Process Model
https://t.co/ytvLA9FbR6
Source Map Explorer
https://t.co/CUA5V0Z89v
AST Explorer
https://t.co/uGyCacLAYQ
Babel Parser
https://t.co/Bys2o2MwU9

Also useful: Electron Security Guide · source-map · Esprima · Acorn · JS Beautifier

08 - BROWSERS AND WEB APPS

Chrome DevTools Protocol
https://t.co/nQ5ZETR8VG
Chrome DevTools
https://t.co/FlRUsqxIQi
Puppeteer
https://t.co/IHJJEOOHWc
Playwright
https://t.co/sapocTF01g

Also useful: OWASP ZAP · HTTP Toolkit · WebSocket King · Firefox DevTools · HAR Specification

09 - FIRMWARE AND DEVICES

Binwalk
https://t.co/duopoSyRQH
EMBA
https://t.co/szbIYqXeA1
FACT Core
https://t.co/bNi2qyNMcf
FirmAE
https://t.co/ElmRiU6pak

Also useful: sasquatch · ubi_reader · Jefferson · Firmware Mod Kit · Firmware Analysis Toolkit · Revenant

10 - WEBASSEMBLY

WABT
https://t.co/wts4HHCdG8
Binaryen
https://t.co/4xhnwZh5DR
wasm-tools
https://t.co/TRFV2gyaIj

Also useful: Wasmtime · Wasmer · wasm-decompile · wasm2c

11 - NETWORK PROTOCOLS

Wireshark
https://t.co/Fv0rUB4Hm8
Scapy
https://t.co/DqAC2ditwk
Zeek
https://t.co/Jb1Jqn25F1
Boofuzz
https://t.co/nyzyjH9XWn

Also useful: tcpdump · PcapPlusPlus · Construct · Netzob · Wireshark Display Filters

12 - FREE COURSES AND PRACTICE

OpenSecurityTraining2 x86-64 Assembly
https://t.co/7OV4XkcoO0
pwn.()college Reverse Engineering
https://t.co/guxhlodUB8
pwn.()college Reversing Hell
https://t.co/Y3M497RwEa
Malware Unicorn RE101
https://t.co/SSwWK3JDGq

Also useful:

RPISEC Modern Binary Exploitation · UMass Reverse Engineering · OpenSecurityTraining Intro x86 · crackmes.()one · Microcorruption · Flare-On · Malware Traffic Analysis · OWASP MASTG · OverTheWire · PicoCTF

THE FASTEST PATH

Install REA

→ investigate one feature
→ collect Evidence
→ trace the complete function
→ write a clean-room specification
→ build an independent implementation
→ compare it against the authorized original
→ keep every failure and unknown visible
Start small:

Notes → Notion → CTF binary → Aegis APK → DX-Ball → firmware, protocols and real applications

Save this. Pick one target. Reverse engineer one feature.

[Cheery @cheery9998]

从零搭建「第二大脑」。

Obsidian 的创始人 kepano。把让 AI 理解 Obsidian 的 Skills 开源了。

GitHub 上有4 个现成项目，几乎把整套流程都准备好了。

① obsidian-skills — kepano

49.8K Stars。

Obsidian 官方生态的 AI Skills，让 Claude 理解 Markdown、双向链接、Canvas 等 Obsidian 原生格式。

② claude-obsidian — AgriciDaniel

15K Stars。

直接提供 Wiki、知识摄取、查询、Lint 和保存命令。

不需要从零写工作流。

③ obsidian-second-brain — eugeniughelbur

4.8K Stars，48 个命令。

支持 Claude Code、Codex、Gemini 等多个 AI Agent。

④ second-brain-starter — coleam00

它不直接提供知识库，而是先采访你，了解你的工作、目标和项目，再生成一套适合你的 9 阶段建设计划。

但真正有意思的，不是这些仓库。

而是 Andrej Karpathy 提出的整个思路：

不要让 AI 每次对话都重新搜索你的知识，而要让它持续编译和维护你的知识。

[Palak @Palakonweb]

Vibecoders thank me later.
https://t.co/t8pOppGk84