Alex Burov

I built Postman for MCP servers. MCP has 97M+ SDK downloads and 10k+ servers. The only dev tool is a basic debugger from Anthropic. So I built Kondukt — test, validate, and scaffold MCP servers from your terminal. 3-second demo 👇

@anhphong_dev @ai_studioxyz Sounds interesting and useful. Test your mcp with kondukt 😏

@AlexBurovMain @ai_studioxyz mailsink - temp email api with mcp server built in. agents receive OTPs and magic links directly, no dom scraping. shipped yesterday

I built Postman for MCP servers. MCP has 97M+ SDK downloads and 10k+ servers. The only dev tool is a basic debugger from Anthropic. So I built Kondukt — test, validate, and scaffold MCP servers from your terminal. 3-second demo 👇

@ai_studioxyz What are you building with MCP?

@AlexBurovMain 10 seconds! Full project compile lightning fast. Needed no CI config tweaks.

@ai_studioxyz I'm glad, thank you for your comment and test.

And one more thing: Kondukt is itself an MCP server. claude mcp add kondukt -- npx kondukt serve Ask Claude Code to validate your MCP server — it'll run the checks itself. GitHub: github.com/kondukt-dev/co… Feedback welcome. #MCP #BuildInPublic

Bonus feature most people miss: kondukt agent-docs . --target claude --stdout Analyzes your codebase, generates CLAUDE.md (or AGENTS.md / GEMINI.md). Static analysis. No LLM calls. Runs offline.

@shcallaway Skill issue fr. U just need more skill and tools for that

OVERRATED: running tons of agents in parallel; working on too many things at once; perpetual context-switching; opening lots of low-quality PRs that may never land. UNDERRATED: using one or two agents at a time; focusing on the task in front of you; thinking deeply; finishing stuff; making your code works in prod.

@Adidotdev Claude solo. Best models, best infrastructure, biggest community of strong devs

Who is winning the AI race ? - Claude - ChatGPT - Codex - Gemini - Grok - Something else

@durov We need Telegram App IDE + Claude Code/Codex for starting singular evolving of telegram

We’ve done more than all our competitors combined — yet it’s not nearly enough. We need to do 10–20× more.

@VadimStrizheus Claude's advantage is really good infra around him. Routines, different terminals, apps, skills, pretty apps and UI. And biggest disadvantage of local LLM is weak PC, because for real LLM's u need a server's power

Kimi K2.6 is literally Opus 4.7 open-source. You can run this on 2 512GB Mac Studios in 2026. Imagine running localized intelligence that's actually good. 3 months ago no one would've believe it. the future is open-source

@budapp Good emulator can solve captcha and be invisible for netwatch for automate operation work

Introducing Bud. The first AI Human Emulator. Bud has a full computer with storage, compute, and memory to build and code, sms and telegram to communicate, a full browser to use, can create/store/edit files, connect and use your tools, learn custom skills, work fully autonomously, and complete any task end to end just like a human. Text the number below or try free at bud [dot] app. Comment for 100k free credits.

@eng_khairallah1 Too long for me. I'm just use superpowers + agent-deck and it's perfect

🚨 Anthropic's own team just showed how to actually use Claude Code properly. 25 minutes. free. from the people who built it. watch the workshop. bookmark it. you've been using Claude for months without knowing the commands. Then read the guide below.

@gitlawb Good job! Do u have vscode plugin or web UI? Not only terminal

while Anthropic quietly removed Claude Code from the $20 Pro plan (now $100+ Max only) we just released OpenClaude v0.6.0 completely free & open-source.

@JamilKhabir396 Hi 👋

Even if you have 0 followers Just like and şey "HI" We follow you

@yifan_zhang_ Yes, only the tops are already using it, and so far it doesn't look like singularity, ASI, AGI, etc. Technology itself needs to be developed

Recursive self-improvement via coding agents is the top priority for all frontier labs.