ChatGPT + Lever: What Actually Connects in 2026
There’s no official Lever app in ChatGPT as of July 2026, and Lever doesn’t ship its own MCP server. That puts it behind the curve in its own category — Greenhouse launched a governed MCP in May 2026 — but it doesn’t mean the two can’t talk. Community MCP servers (stefanoamorelli/lever-mcp exposes 59 tools over Streamable HTTP), hosted bridges (Apideck, Zapier MCP), and Lever’s own REST API via custom GPT Actions all work today. Every path shares the same shape: your Lever API key does the talking, ChatGPT answers in a session you’re driving, and between chats nothing watches candidates move through your pipeline.
Here’s what a ChatGPT Lever integration actually looks like today, how to set it up, and what to use when you want recruiting work that runs without you.
What ChatGPT can actually do with Lever
- Search candidates conversationally. “Find candidates in the Product Designer pipeline who’ve reached onsite but haven’t been scheduled” — answered from live Lever data via an MCP connection.
- Summarize pipeline health. Openings by stage, time-in-stage outliers, sources that actually convert — the questions that otherwise mean exporting to a spreadsheet.
- Pull candidate context before a call. Resume highlights, stage history, and past feedback for a candidate, assembled into a briefing in one prompt.
- Draft with real pipeline data. Rejection notes, interview confirmations, and hiring-manager updates written against actual candidate records — drafts you copy out, since posting back depends on your bridge’s write tools.
- Take pipeline actions, if your bridge allows writes. Community and hosted MCP servers expose actions like adding notes or moving stages — enable write tools deliberately.
- Run inside agent sessions. With ChatGPT Work (launched July 9, 2026), you can @-mention connected apps and let an agent work across Lever and the rest of your stack in a long, metered run — a quarter’s recruiting retro, say. Still a run you start.
How to set it up
- Generate a Lever API key from Settings → Integrations and API — this needs admin access, and the key’s scopes decide everything the connector can see and do. Candidate data is sensitive; scope read-only unless you have a reason not to.
- Pick a bridge: a hosted MCP server like Apideck or Zapier MCP, or self-host a community server like lever-mcp.
- In ChatGPT, open Settings → Connectors → Add custom connector (Pro, Business, and Enterprise plans) and point it at the MCP endpoint — or wire the Lever API into a custom GPT via Actions.
- Ask something verifiable (“how many active candidates are in the Sales AE pipeline?”) and check it against Lever before trusting bigger questions.
The limits that actually matter
- No official server means you carry the risk. Community MCP servers are unaffiliated with Lever — audit what they log and where they run before handing over a key that can read your entire candidate database.
- One admin key, everyone’s data. Lever API keys aren’t per-user the way Greenhouse MCP’s governed access is. A key created for a connector sees whatever its scopes allow, regardless of who’s typing the prompt. Treat connector access as an org decision, not a personal one.
- It doesn’t run on triggers. There’s no “when a candidate completes the onsite, chase panel feedback” or “when an offer is extended, prep onboarding.” ChatGPT queries Lever when you prompt it — it never fires on a Lever event.
- Session-bound, even in agent mode. ChatGPT Work runs are long and autonomous but manually started and metered against your plan’s allowance — an errand, not a standing watch on your pipeline.
If you want Lever work that runs on its own: Carly
The moment you want something to happen off a hiring event — panel feedback gets chased the morning after an onsite, hiring managers get a pipeline digest every Monday, a candidate sitting in offer stage for three days triggers a nudge — you’ve crossed past what a session-bound connector is for.
That’s where Carly fits. Carly is an AI executive assistant that acts on triggers across your whole stack, set up by conversation instead of code:
- Fires on events and schedules, 24/7, in the cloud. Stage change, completed interview, Monday 8am — Carly acts without a chat open.
- No-code setup. Tell Carly “every Monday, email each hiring manager a digest of their open roles in Lever” in plain English; it interviews you and builds the workflow.
- Connects recruiting to the rest of your work — Lever data flowing into email, calendars, Slack, and spreadsheets in one flow.
- Actually sends — drafts and sends email across Gmail and Outlook, schedules interviews, manages tasks.
- Connects to anything — Carly integrates with Lever natively, alongside 200+ other native integrations, plus any tool via your own API key.
AI agents start at $35/month, and steps in a workflow that don’t use AI run free and unlimited. See integrations for the full list.
ChatGPT vs Carly
| ChatGPT (via third-party MCP) | Carly | |
|---|---|---|
| Candidate search & pipeline analysis | Yes | Yes |
| Pre-interview candidate briefings | Yes, when you ask | Yes, automatically before each interview |
| Native Lever integration | No (third-party bridge + your API key) | Yes |
| Monday pipeline digest, unprompted | No | Yes, on a schedule |
| Reacts to a stage change by itself | No | Yes, on any trigger |
| Runs without a session open | No (agent runs are started + metered) | Yes (cloud, 24/7) |
| Emails panels and hiring managers | No | Yes (Gmail + Outlook) |
| Setup | Choose a bridge, manage keys | Describe it in plain English |
| Pricing | Paid ChatGPT plan + connector | AI agents from $35/mo |
ChatGPT with a Lever bridge is a recruiting analyst you question in a chat. Carly is an assistant that acts on your pipeline while you’re doing something else.
Frequently Asked Questions
Does ChatGPT work with Lever?
Not officially — there’s no Lever app in ChatGPT’s directory and Lever doesn’t publish its own MCP server as of July 2026. You can connect them through community or hosted MCP servers added as custom connectors, or through Lever’s REST API with custom GPT Actions.
Is there a Lever MCP server?
Not from Lever itself. Community servers like stefanoamorelli/lever-mcp (59 tools) and hosted options from Apideck and Zapier fill the gap. They’re unaffiliated with Lever, so vet where your candidate data flows before connecting one.
Can ChatGPT react to a Lever event automatically?
No. ChatGPT queries Lever inside a session you start — it doesn’t watch for stage changes, completed interviews, or aging offers. For “when X happens in Lever, do Y across my stack,” you need a trigger-based assistant like Carly, which integrates with Lever natively.
How do I connect ChatGPT to Lever?
Generate a Lever API key (admin access required), pick an MCP bridge — hosted like Apideck or Zapier MCP, or self-hosted — then add it in ChatGPT under Settings → Connectors → Add custom connector. Scope the key read-only unless your workflow genuinely needs writes.
More: ChatGPT email assistant · ChatGPT personal assistant · ChatGPT MCP · Can ChatGPT send emails · Claude + Lever · Best AI agents for productivity
Ready to automate your busywork?
Carly schedules, researches, and briefs you—so you can focus on what matters.
See what people say
"Before Carly, I relied on a Calendly link, but the whole process felt impersonal and not very professional. Carly changed that by handling all the back-and-forth, so I'm no longer stuck in endless email threads trying to line up schedules.
Now Carly reaches out to candidates, shares my real-time availability, lets them pick a slot, then sends a Zoom link and drops it straight into my calendar. She sends reminders to both of us before each call, which has significantly reduced no-shows and last-minute confusion.
On top of scheduling, Carly acts like a full executive assistant, sending me my schedule the night before so I can prepare for each call. It reminds me of the old x.ai assistant, but Carly is noticeably smarter, faster, and better suited to my healthcare recruitment business."
