Leo Li

Google Flights for agents — but for APIs. (10+ domains, 100+ APIs) We’re building the unified layer for agentic APIs.

I'm verifying my builder profile on FeatDrop(@featdrop_team), the public changelog for product builders to share updates. Verification code is ULW-770. Follow me at featdrop.com/u/leo

@hanghuang_ Love the story!

After 6 applications and 6 rejection emails, we finally got into Y Combinator. Yes, read that one more time: 6 applications. 6 rejections. We turned a rejection into an admission offer. For a long time, every rejection led to the same question: "Do we pivot, or keep going?" We didn't think much of the first few rejections. Our reaction was mostly just: okay, back to building, apply again next time. Honestly, the hardest one was the 5th rejection because we felt so close. It was the first time we got an interview. We believed we had a real shot. But in the end, we got rejected… again. Looking back, the decision was fair. We were only doing around $300/month, and YC didn’t see a clear path to building a billion-dollar company through enterprise. So we stopped guessing and started listening. We did 20+ user interviews and realized something important: the people who really loved InsForge were not big enterprises. They were AI-native small teams and startups. That fundamentally changed how we saw the company. We clarified who the product was actually for, doubled down on what was working, and kept building in public on X and LinkedIn. We grew from 2,300 to 4,000+ databases in 2 months. Then we applied again. Our second interview with YC. We really thought this would be the one. But once again, we were rejected. That was the moment the question we had been asking ourselves after every rejection finally changed. No longer: “Do we pivot?” Instead: “How do we execute so well that the need for this product becomes impossible to ignore?” After 30 days of hell, we launched @InsForge_dev Launch Week 1. And it took off. Like, really took off! → 1.5M+ views on X → #1 on Product Hunt → #1 on GitHub Trending → 3K+ GitHub stars in one week But here's the craziest part: after rejecting us, YC changed their mind. Here was our second chance. We got an email from general partner Andrew Miklas (@amiklas), congratulating us on our launch and asking us to meet one more time. We figured it would be another tough interview. But the meeting was in two hours. No time to prepare. We were so nervous up until the very end. When we finally hopped on the call, he just said, “You guys have made huge progress. I want to work with you. Do you want to do YC?” WTF????????? Tony (@tonychang430) and I looked at each other. We were so shocked, we didn't even know what to say. Of course, the answer was yes. This is when we learned: Execute so well that your company becomes impossible to reject. Every rejection forced us to clarify our vision. The last one forced us to prove it. Next stop: YC P26!! @ycombinator 🥳 ( Read the full story below ⬇️ )

Agents don’t want API keys. They want outcomes. We just launched: agentdiscuss.com/agentic-api → 100+ APIs (working on) → pay per request → no signup, no credentials Agents describe the task we route + execute with the best provider

@adisingh lets go

We just launched task runs on AgentDiscuss — to measure how coding agents actually use APIs. We ran 4 tasks across email APIs. Each one breaks in a different place.

@VadimStrizheus @agentdiscuss

what’s the best agentic tool out there? 1. OpenClaw 2. Hermes Agent 3. PaperClip (Hermes Agent looks promising) 🤷‍♂️

@shivsakhuja @agentmail @tryagentphone and also @agentdiscuss :P

Lots of companies are now building primitives for an economy where AI agents are the primary users instead of humans. They're betting on an economy of AI coworkers. 1. AgentMail (@agentmail): so agents can have email accounts 2. AgentPhone (@tryagentphone): so agents can have phone numbers 3. Kapso (@andresmatte): so agents can have WhatsApp phone numbers 4. Daytona (@daytonaio) / E2B (@e2b): so agents can have their own computers 5. Browserbase (@browserbase) / Browser Use (@browser_use) / Hyperbrowser (@hyperbrowser): so agents can use web browsers 6. Firecrawl (@firecrawl): so agents can crawl the web without a browser 7. Mem0 (@mem0ai): so agents can remember things 8. Kite (@GoKiteAI) / Sponge (@PayspongeLabs) : so agents can pay for things. 9. Composio (@composio): so agents can use your SaaS tools 10. Orthogonal (@orthogonal_sh) so agents can access APIs easily 11. ElevenLabs (@ElevenLabs) / Vapi (@Vapi_AI) so agents can have a voice 12. Sixtyfour (@sixtyfourai) so agents can search for people and companies. 13. Exa (@ExaAILabs): so agents can search the web (Google doesn’t work for agents) If you stitch all of these together, you get a digital coworker that looks more human than AI.

@sv2l_ @agentdiscuss just DM'ed you

@ideapoet @agentdiscuss I can’t if you don’t follow me...

I am verifying my AI agent for @agentdiscuss to see what products agents prefer. Here is my code: aurora-BA9E

Agents are already deciding which dev tools win. We ran 8 tasks across 4 coding agents. Results are not what you'd expect: SendGrid > Resend Hono > Express Stripe > LemonSqueezy Prisma > Supabase Different rules now. → agentdiscuss.com/?kind=task

@sv2l_ @agentdiscuss Thanks! Feel free to DM!

@ideapoet @agentdiscuss hi Leo, congratulations for what you did with AgenDiscuss. I would like to connect with you to make you a proposal, let me DM you

@evertjr Congrats on the launch! Might be interesting to also share it on @agentdiscuss — agents there actually evaluate and vote on products.

Oi gente! Hoje venho apresentar o Maestri, o melhor aplicativo para orquestrar agentes de IA. Se torne um dev 100x! Criei esse app para aumentar minha produtividade e vim compartilhar com vcs. Trabalhe em um inovador canvas 2D onde você e seus agentes colaboram em harmonia.

Task runs. See how coding agents pick tools. → agentdiscuss.com/?kind=task

We just shipped some updates to AgentDiscuss: • You can now claim your product → add an official agent-ready spec (MD) • New product summaries → see how agents evaluate your product • A redesigned browsing experience → easier for both humans & agents Still early — would love feedback from builders.

I am verifying my AI agent for @agentdiscuss to see what products agents prefer. Here is my code: fjord-BE4F

@andrewchen @agentdiscuss is helping CLI or agent tools to discover default agent tools

Web 1.0 came with new channels: - email, search, link sharing, etc Web 2.0 too: - feeds, creators, viral invites, etc Mobile: - app stores, SMS invites, vertical vid, mobile ads What about AI? I’ve been complaining that AI hasn’t come with much. But we’re seeing a big growth channel opening now: Products that are built as APIs/CLIs that can be pulled into new projects by Codex/Claude on the fly Maybe the “AI-native hotel app” doesn’t mean a mobile booking app with an AI chat panel. It means a CLI that can book a hotel for you, that an AI agent can pull into a bespoke answer or project or into code. Bolting on an AI chat panel is this generation’s weak form of AI. Maybe the full reinvention involves making it agent-first not human-first and once you start looking at it that way, a lot of existing products suddenly feel mis-specified. they’re built as destinations, but agents don’t want destinations. they want capabilities. composable, callable, reliable capabilities. So instead of “go to Expedia” or “open the app,” the future interaction is more like: an agent assembles a workflow on the fly. it pulls a flight search tool, a hotel booking tool, maybe a weather model, maybe even your personal preference graph. none of these are full products in the traditional sense. they’re more like endpoints with taste and state. This flips distribution completely. historically you win by owning the surface area. seo, app store ranking, homepage traffic. in an agent world, you win by being the default callable primitive. the thing that shows up again and again in agent-generated plans because it works, has clean interfaces, and returns structured outputs. distribution shifts from “top of funnel” to “top of call stack.” And the crazy part is this might actually compress product surface area dramatically. the best products might look more like tight, extremely well-designed CLIs with opinionated defaults rather than sprawling UIs. almost like the stripe api moment, but for everything. imagine if every vertical had a “stripe-level” primitive that agents preferentially use. there’s also a weird inversion of brand here. humans used to choose brands. now agents will. so the brand becomes partially machine-legible. reliability, latency, error rates, schema clarity. you can almost imagine “agent seo” where the ranking factors are things like success rate across thousands of agent runs, or how easy your tool is to integrate in a chain-of-thought execution loop. This also suggests a new kind of moat. not just data or network effects, but integration depth with agent ecosystems. if claude or codex or openclaw learns that your tool is the safest way to accomplish X, it gets baked into prompts, templates, maybe even fine-tunes. you become a default. and defaults, historically, are insanely sticky. The contrarian take is that most current “AI features” are a local maximum. chat panels, copilots, assistants. they’re transitional. the real end state might look closer to invisible infrastructure that agents orchestrate. the ui is just a debug layer for humans to peek into what the agents are doing. so maybe the new growth channels for ai look like: - being callable - being composable - being reliable at scale in agent loops - being embedded in agent templates and workflows - being the default primitive in a given domain and if that’s right, then the question for any new product isn’t “what’s the ui” or even “what’s the killer feature.” it’s “what’s the minimal, highest-leverage capability we can expose such that agents will repeatedly choose us when building something new.”

It works like a product exploration layer for OpenClaw or company agents since agents are debating different dev tools from different dimensions. I think the goal is to understand more on how agent makes choices and how to build better agent tools! Definitely send it to AD and give a try.

@ideapoet @runchao_han Thanks for the mention! AgentDiscuss sounds interesting — agents discussing their own products is very on-brand for what we're building. How does it work?

ClawPier v0.3.0 is out! 🦞 New: ClawHub skill browser, health checks w/ auto-restart, port mapping presets, dark mode, and now cross-platform (macOS/Linux/Windows). macOS builds are signed & notarised. github.com/SebastianElvis…

@runchao_han Thanks. Would also be interested in learning more about ClawPier. Any competing options now? Defintely sending it to AD to tell more agents!

@ideapoet Nice, thanks for the information!