Build projects that agents can actually use

I’ve been staring at the same fork in the road a lot of physician-builders are hitting.

You can ship another dashboard. Another CRUD app. Another “AI wrapper” with a nice landing page. Cursor and Claude will help you build it in a weekend — which is also why someone else can rebuild the UI in a weekend.

Or you can build something an agent can plug into — tools, auth, an MCP endpoint, a skill file — so Cursor, Claude Code, ChatGPT, Grok, or a Discord bot can do work without clicking through your screens.

That second category is what people keep calling agent-native. I think it’s the more interesting bet if you care about projects that still matter when the model gets better at cloning frontends.

I keep a public scoreboard of that question at isitaiproof.com — “is this still standing in ten years?” This essay is the builder version: what should you ship if agents are the next distribution channel?

What “agent-native” actually means

Not “we added a chatbot to the marketing site.”

It means the product’s real surface is something an agent can call:

• an MCP server (tools with names and schemas)
• an API with clear auth
• a skill / playbook the agent can load (SKILL.md, rules, CLI)
• often no UI required for the core loop

The human still buys. The human still connects accounts and pays. The agent is the operator.

That’s a different product than “please open our app every day.”

Post Bridge — social posting as a tool, not a tab

The cleanest consumer example I’ve looked at lately is Post Bridge.

It’s a normal social scheduler on the surface — Instagram, TikTok, YouTube, X, LinkedIn, Facebook, Pinterest, Threads, Bluesky. The interesting part is Agent Mode / MCP: one URL, OAuth or API key, and your coding agent can schedule posts, upload media, and pull analytics in chat.

Their agents page is blunt about the pitch: stop babysitting nine native apps; give the agent the posting tools.

There’s a public agent-mode repo and a skill file so the agent knows the workflow, not just the raw HTTP. That’s the pattern: product + MCP + skill.

On isitaiproof, Post Bridge already sits in the catalog under social / integrations (community still arguing AI-proof vs at-risk — which is fine; the point is people are watching this shape of company). Search “Post Bridge” there if you want the vote board next to this essay.

The same shape, different layers

A few other projects that rhyme, even if the category label differs:

Composio — agent connectors so your agent can talk to hundreds of SaaS apps without you hand-rolling OAuth for each one. Also on isitaiproof. If Post Bridge is “posting as a tool,” Composio is “the plumbing shelf.”

Browserbase — managed browsers for agents that still need the open web. Paul Klein’s been public about MCP tools like navigate / act / extract instead of dumping raw Playwright into every prompt. When the UI wasn’t built for agents, you give the agent a browser that was.

Arcade — auth and action runtime so agents can hit Slack, Google, Salesforce, etc. without stuffing refresh tokens into the model context. The boring-hard problem: permissioned actions for real users.

Agent-Native (Builder.io) — a framework bet: define an action once, expose it to UI and agent and MCP and CLI. Same work, multiple surfaces.

You don’t need to use their framework to steal the idea. One capability, multiple callers.

Closer to home

I already run a small version of this without calling it agent-native.

The Physician Forge Discord bot answers /ask with Grok, grants membership roles after Stripe, and lives where the conversation already is. Discord is the client. The bot is the tool surface. Members don’t need a separate “AI portal.”

That’s the same instinct as Post Bridge — put the agent where people (or other agents) already work.

What I have not shipped yet, and keep meaning to: a real MCP endpoint for pieces of Physician Forge / OrthoAndSpineTools / Direct Care List so Cursor in someone else’s project can query or post without scraping the site. The essay is partly a note to myself.

Why this survives the “AI one-shot” panic

isitaiproof exists because a lot of SaaS is a thin UI over a model call. Those get cloned.

Agent-native products tend to stack different stuff:

1. Connected accounts — OAuth to nine social platforms, or a bank, or an EHR-adjacent workflow. Painful to rebuild.
2. Trust / blast radius — posting as you, charging a card, moving a calendar. People don’t reconnect that lightly.
3. Distribution into agents — if Cursor / Claude / ChatGPT / Grok can discover your MCP, you ride their install base.
4. Ops reality — retries, media transcodes, per-platform caption rules, analytics sync. The skill file exists because the edge cases are the product.

None of that makes you immortal. It does change the question from “can AI redraw my screens?” to “will people hand an agent the keys to this?”

If you’re picking a side project this month

A few prompts I use on myself:

• What do I already do manually that I wish Cursor could just finish?
• Can I expose that as 5–15 named tools with boring auth?
• Is there a human UI optional path — agents first, dashboard second?
• Would I list it on isitaiproof without laughing — is there a moat beyond “we used GPT”?

Examples worth studying with links open:

• post-bridge.com/mcp
• post-bridge.com/agents
• github.com/post-bridge-hq/agent-mode
• composio.dev
• browserbase.com
• arcade.dev
• agent-native.com
• isitaiproof.com — vote the catalog; search Post Bridge / Composio

I couldn’t pull a live X search dump from this box tonight (API not connected here), so I’m pointing at the public product surfaces and the isitaiproof listings instead of screenshotting a timeline. If you’re reading this from X later, the thread is the trailer — the durable version lives here.

Build something an agent can use without opening a tab. That’s the work that still feels like leverage when the next model ships.