Bullhorn MCP Server
आपके Bullhorn ATS में उम्मीदवार, नोट्स और सबमिशन।
होस्ट किया गया MCP सर्वर
npx add-mcp 'https://cotlet.dev/mcp/v1/bullhorn'Claude Code, Codex, Cursor और अन्य में इंस्टॉल होता है
दस्तावेज़
Cotlet · Integrations · How it works · Security · For developers · For automation agencies · Blog · API docs · Help center · Changelog · Contact · Pricing · Sign up
Bullhorn API and MCP server: Let your AI agent keep your Bullhorn desk up to date.
Bullhorn has no MCP server (checked 27 Oct 2026). It has a REST API, but customers get keys only through a support ticket. Cotlet runs 9 Bullhorn tools hosted, over MCP for agents and REST for scripts, in the login you already use.
Candidates, notes and submissions in your Bullhorn ATS. Use it in ChatGPT, Claude or any AI agent over MCP, or in n8n, Make and scripts over REST.
- Free 7-day trial, no card
- Unofficial, not affiliated with Bullhorn
- Page: https://cotlet.dev/integrations/bullhorn-api-mcp-server
Cotlet is an independent service. It is not affiliated with, endorsed by or sponsored by Bullhorn. Bullhorn is a trademark of its owner. If Bullhorn's own integrations cover what you need, use them first.
What can you automate in Bullhorn?
- Log screening calls as candidate notes
- Match a new job to your database
- Submit your picks to the job owner
- Clear submissions stuck at Submitted
- Count each job's pipeline every Monday
- Find past candidates worth a call
Does Bullhorn have an MCP server or an API?
- Vendors need a paid API Access contract: an annual fee paid upfront, 200,000 calls a month and a security assessment. Bullhorn does not let connector or middleware companies build on it.
- There is no Bullhorn app on Zapier and no official n8n node (both pages answer 404).
- Bullhorn's own AI, Amplify Chat, works over live data inside Bullhorn on the Pro and Max plans. No outside agent can call it.
| What you need | Bullhorn REST API | Amplify Chat | Cotlet |
|---|---|---|---|
| Search candidates and read notes | Yes, with a ticket key | Yes, inside Bullhorn | Yes |
| Add notes and submissions | Yes, as API calls | Not documented | Yes, after your OK |
| The app's Find Matching Candidates rule | No endpoint | AI search and match | Same rule as the app |
| Use from your own AI agent | Only with written permission | No | Yes, any MCP client |
| Bullhorn plan | Any, key by ticket | Pro or Max, by quote | The plan you have |
Checked 27 Oct 2026. Sources: Bullhorn API Fair Use Policy, Bullhorn API Access FAQs, Bullhorn developer FAQ: getting a REST key, Bullhorn pricing
How does Cotlet connect to Bullhorn?
- Sign in to Bullhorn. Use your own account in its Cotlet Cloud Browser.
- Choose MCP or REST. Connect your agent or create a Cotlet API key.
- Run your workflow. Ask your agent or send an HTTP request.
We never keep your Bullhorn password, and no note, submission or status change is saved before you approve it.
Connect Bullhorn to your agent or workflow
Use MCP with your AI agent, or HTTP requests in n8n, Make, Zapier and scripts. Follow the MCP setup below or jump to the REST guide.
- Log in to Bullhorn in a Cloud Browser. Create a Cotlet account, open Bullhorn in its Cloud Browser and log in with your own Bullhorn username, or through your Identity Provider if your agency uses SSO. Then press Enable for agents.
- Add the Cotlet address. Paste https://cotlet.dev/mcp/v1 into the MCP server or connector field of your agent.
- Approve the connection. Chat apps show a Cotlet sign-in page. Code editors, n8n and scripts send an API key you create in the console.
Bullhorn shares https://cotlet.dev/mcp/v1 with every other app in your Cotlet account. If you enable Bullhorn after the agent is connected, refresh the connector to load the 9 tools.
Claude:
Settings > Connectors > Add custom connector
URL: https://cotlet.dev/mcp/v1/connect
Claude Code:
claude mcp add --transport http cotlet \
https://cotlet.dev/mcp/v1 \
--header "Authorization: Bearer $COTLET_API_KEY"
ChatGPT:
Settings > Apps & Connectors > Create
MCP server URL: https://cotlet.dev/mcp/v1/connect
Authentication: OAuth
Cursor:
// ~/.cursor/mcp.json
{
"mcpServers": {
"cotlet": {
"url": "https://cotlet.dev/mcp/v1",
"headers": { "Authorization": "Bearer ${env:COTLET_API_KEY}" }
}
}
}
Codex:
# ~/.codex/config.toml
[mcp_servers.cotlet]
url = "https://cotlet.dev/mcp/v1"
bearer_token_env_var = "COTLET_API_KEY"
n8n:
MCP Client Tool (inside an AI Agent)
Transport: HTTP Streamable
URL: https://cotlet.dev/mcp/v1
Auth: Bearer, <your key>
Tool: bullhorn_read
Make, Zapier:
Make: MCP Client > Call a tool
Zapier: MCP Client by Zapier > Run Tool
Server URL: https://cotlet.dev/mcp/v1
Token: <your key>
Tool: bullhorn_write
How to use the Bullhorn REST API
Create your Cotlet API key
Create a Cotlet API key with access to Bullhorn and its read operations. Store it as COTLET_API_KEY in your environment or automation platform's credential store.
Connect Bullhorn in Cotlet and enable it for agents. Account access: Your own Bullhorn ATS user (the same username and password you use today; if your company turned on email access codes, you enter the code yourself).
Make your first Bullhorn API request
REST:
curl "https://cotlet.dev/api/v1/bullhorn/candidates/search?keywords=dbt&limit=5" \
-H "Authorization: Bearer $COTLET_API_KEY"
Read candidates and apply the operation limits
A successful REST call puts the tool output in result. This illustrative response excerpt omits call IDs, messages and credit details.
Illustrative response excerpt:
{
"status": "success",
"app": "bullhorn",
"tool": "search_candidates",
"result": {
"total": 31,
"candidates": [
{
"candidate_id": 4821907,
"name": "Priya Raman",
"occupation": "Data Analyst",
"company": "Brightpath Retail",
"status": "Active",
"submission_count": 4
},
{
"candidate_id": 4830112,
"name": "Marcus Lee",
"occupation": "BI Lead",
"company": "Harbor Logistics",
"status": "Active",
"submission_count": 1
}
],
"next_cursor": null
}
}
Searches the way the candidate list keyword search does: Candidate Name, Occupation, Current Company, Resume, Work History and File Attachments. Custom fields are not searched.
Several keywords combine with and by default; match any uses or, and exclude_keywords removes candidates that contain them. Returns one page of candidates; pass next_cursor back as cursor for more.
Use candidate_id in get_candidate and add_note.
limit: Candidates per page.
cursor: Next page cursor from the previous call.
Replace the illustrative dbt keyword with your recruiting search. Preserve any account permissions required by the reference.
Keep the same selection when continuing a list. Compare changed record values as well as identifiers when refreshing your destination.
Bullhorn workflows for agents and automations
| Tool | Connect through | Example workflow |
|---|---|---|
| n8n | HTTP Request | One candidate record with status, current company and the jobs they were submitted to. Use get_candidate in your configured team report. |
| Make | HTTP > Make a request | Notes on a candidate's Notes tab with Note Action, author and date, newest first. Use list_candidate_notes in your configured team report. |
| Python or Node.js | REST API | Submissions on a job by stage, with candidate, status and date. Use list_job_submissions in your configured team report. |
Send Authorization: Bearer <your Cotlet API key> in request headers. You configure the schedule, destination and record comparisons.
Cotlet reads Bullhorn; your workflow steps write to your chosen destination. These HTTP workflows do not require a native Cotlet connector or a prebuilt template.
Python and Node.js can use the same HTTP requests without an AI agent. Keep access keys in credentials, outside shared workflow exports.
The API and MCP reference lists operations and inputs. The REST API guide explains authentication, errors and approvals.
Bullhorn API endpoints and MCP tools
Candidates and notes
Search reads name, occupation, company, resume, work history and attachments, never custom fields. A name typed after @ in a note emails that Bullhorn user.
- Search candidates (
search_candidates, read): Candidates matching keywords in name, occupation, company, resume, work history and attachments. - Get candidate (
get_candidate, read): One candidate record with status, current company and the jobs they were submitted to. - List candidate notes (
list_candidate_notes, read): Notes on a candidate's Notes tab with Note Action, author and date, newest first. - Add candidate note (
add_note, write, asks first): Adds a note with a Note Action to a candidate's Notes tab.
Jobs and matching
Matching follows the app: up to 10 keywords from a description over 250 words and a 20 mile radius around the job zip code, unless you pass your own.
- Get job (
get_job, read): One job record with title, client, owner, location and submission count. - Find matching candidates (
find_matching_candidates, read): Candidates that match a job by its description and location, like Find Matching Candidates.
Submissions
A new submission lands with the job owner as Submitted. Moving another recruiter's candidate needs the Change Any Candidate Submission Status entitlement.
- List job submissions (
list_job_submissions, read): Submissions on a job by stage, with candidate, status and date. - Submit candidates to a job (
add_submission, write, asks first): Submits one or more candidates to a job for the job owner to review. - Update submission status (
update_submission_status, write, asks first): Moves one submission to another status in your hiring workflow.
Inputs, REST path and example response of every tool: https://cotlet.dev/integrations/bullhorn-api-mcp-server#tools
Who can see my Bullhorn session, and how do I revoke it?
- What we store. Your logged-in Bullhorn profile, AES-256-GCM encrypted, in Finland. Not the password.
- What your agent gets. 9 Bullhorn tools. No cookie, no username.
- What the session sees. The records your Bullhorn user can open, with its entitlements.
- What no tool touches. Placements, pay and bill, deletes, admin settings and system settings.
- If Bullhorn logs you out. You log in again yourself, with your own MFA code or SSO.
- How to cut access. Disable for agents, reset the browser, or change your Bullhorn password.
After its cyber incident of 17 Oct 2026, Bullhorn made every user without SSO reset the password (The Register, 20 Oct 2026). A reset like that ends the Cloud Browser session too, and the next call asks you to log in again. Our site has A+ on Qualys SSL Labs, securityheaders.com and Mozilla Observatory (live results).
Approvals, undo and mistakes
- Reads the job. Title, client and owner, so it matches the job you meant.
- Runs the match. Find Matching Candidates, then the top records, already submitted ones left out.
- Waits for your OK. The shortlist and the comment sit in your Cotlet console.
- Submits and reports. One submission per pick, each with its new ID.
| Action | Undo |
|---|---|
| Add a note | No tool removes it; anyone tagged with @ already has the email |
| Submit a candidate | Delete it from the job's Submissions tab, as Bullhorn advises for a submission made by mistake |
| Change a submission status | Set the old status back; the row stays |
All 3 write tools ask first: add_note, add_submission and update_submission_status, because a tagged note emails a colleague and a submission lands with the job owner. The 6 reads run at once. For a weekly cleanup you already trust, you can switch the wait off per tool.
Bullhorn lets a recruiter submit several candidates at once only if the agency's policies allow it. Without that, submit one candidate per call.
MCP call:
{
"name": "add_submission",
"arguments": {
"job_id": 77314,
"candidate_ids": [
4821907,
4830112
],
"comments": "Both have dbt and Snowflake and can start in November."
}
}
REST:
curl -X POST https://cotlet.dev/api/v1/bullhorn/jobs/77314/submissions \
-H "Authorization: Bearer $COTLET_API_KEY" \
-H "Content-Type: application/json" \
-d '{"candidate_ids":[4821907,4830112],"comments":"Both have dbt and Snowflake and can start in November."}'
Response:
{
"submissions": [
{
"submission_id": 905601,
"candidate_id": 4821907,
"status": "Submitted"
},
{
"submission_id": 905602,
"candidate_id": 4830112,
"status": "Submitted"
}
]
}
A script or an n8n flow calls the same tools over REST at /api/v1/bullhorn, with one key and one price, for example a Monday pipeline count mailed to each recruiter. A status your account does not have returns an error, and errors are free. Each tool row above shows its REST path.
How much does the Bullhorn MCP server cost?
About $61 a month for an agency with 5 recruiters and 15 open jobs, 10,560 credits. No API contract; Bullhorn Starter is $99 and Core $165 a user a month.
| Job | Calls | Credits |
|---|---|---|
| Log a screening call as a note (find, check, add) | 3 × 160 | 4,800 |
| Match a new job, read 5 records, submit picks | 8 × 12 | 960 |
| Stale submissions on 15 open jobs, weekly | 45 × 4 | 1,800 |
| Pipeline count for 15 jobs, weekly | 30 × 4 | 1,200 |
| Find past candidates to re-engage | 15 × 2 | 300 |
| Cloud Browser, 30 days | 1,500 | |
| Total | 10,560 |
- Free 7-day trial, no card
- 10 credits per successful call, MCP or REST
- 50 credits a day for the Cloud Browser
- No API Access fee, no support ticket
The month is our example: 5 recruiters log about 160 screening calls, match 12 new jobs and check 15 open jobs each week. Bullhorn publishes no activity figures; its Starter plan fits 1 to 2 users and Core starts at 3. Dollars at the $29 pack rate.
Bullhorn MCP server on GitHub, Zapier or hosted: which to use
| Option | What it does | Where it stops |
|---|---|---|
| osherai/bullhorn-mcp-python | 6 read-only tools for jobs and candidates, Lucene and query search. MIT, 1 star, last push 15 Jan 2026. | Needs REST keys from a ticket; no notes or submissions. |
| beaufromit/bullhorn-mcp | Adds add_note and a shortlist-to-job workflow. MIT, 0 stars, last push 6 Oct 2026. | Same keys; you host it. |
| CaseyTTS/bullhorn-mcp-agent | Writes with a preview, then a confirm step, plus resume upload. 0 stars, last push 8 Oct 2026. | Same keys; no standard license. |
| StackOne | Hosted connector over MCP with 47 actions, including notes and job submissions. | Sold to software vendors; price by quote. |
| Knit | Hosted Bullhorn MCP server with reads and writes, OAuth handled. | Needs Bullhorn API access. |
| CData | Bullhorn CRM MCP server on its JDBC driver. | Read-only; paid driver. |
All six need API credentials, which fall under the Fair Use Policy rule on MCPs and outside AI. If Bullhorn gives you written permission and keys, an API server is the cleaner route. If you have neither and want the agent to work your desk, Cotlet works in your login.
Who runs Cotlet
Cotlet belongs to Take Group sp. z o.o., a software company registered in Poland, EU (VAT PL6342984273) and trading since 2019. Its founders are Bart Krawczyk (Co-founder, CTO) and Jake Slugocki (Co-founder, SEO, sales and marketing). Ask about the Bullhorn tools at contact@cotlet.dev.
Bullhorn API and MCP: questions
How do I get access?
Sign up free at https://cotlet.dev, open Bullhorn under Integrations in the console and log in to Bullhorn in its Cloud Browser. The trial starts when you sign up and asks for no card.
How do I connect Bullhorn to ChatGPT or Claude?
Sign up at cotlet.dev, log in to Bullhorn once in its Cloud Browser and add Cotlet to ChatGPT or Claude, following the steps in your console. No API key is needed. ChatGPT and Claude then work in your Bullhorn account through Cotlet, in their phone apps too.
What can ChatGPT do in Bullhorn?
With Cotlet, ChatGPT, Claude and other AI agents handle jobs like these in Bullhorn: log screening notes on candidate records; find and submit matching candidates for a job; clean up stale submissions. You ask in plain words, and anything that reaches other people waits for your OK first.
Can my agency get a Bullhorn API key for its own agent?
Bullhorn sells API access as a partner contract to technology vendors, with an upfront annual fee and a security assessment. Bullhorn also says customers do not get their key if they plan to hand it to a vendor without a contract. Cotlet does not use that API: your agent works in your own Bullhorn login in the Cloud Browser.
Which Bullhorn role do I need?
Your own user, with the rights it already has. Your agent can only open records and do actions that your user can do in Bullhorn. Changing the status of a candidate that is not yours needs the Change Any Candidate Submission Status entitlement, and a call without it fails.
Does it work with SSO and MFA?
Yes. Choose Log in via Identity Provider in the Cloud Browser, or type the code Bullhorn sends you. You do both yourself; we never enter an MFA code for you.
Can I use Bullhorn's own AI instead?
Amplify Chat answers questions over your Bullhorn data on the Pro and Max plans, inside Bullhorn. If you are on those plans and only need answers in the app, it may be enough. Cotlet is for an agent you already run, such as Claude, ChatGPT or n8n.
Is the MCP server different from the REST API?
Only in how you call it. Both expose the same 9 tools, inputs, answers, key and price. Agents use MCP; scripts use REST.
Is Cotlet affiliated with Bullhorn?
No. Cotlet is independent and not affiliated with Bullhorn, Inc.
Can I use the Bullhorn API without an AI agent?
Yes. Call Cotlet's REST API from a script or an HTTP step in n8n, Make or Zapier with your Cotlet key. MCP is the alternative for AI agents; you do not need to connect an agent to use REST.
Give your AI agents your signed-in browser.
Product: Integrations · How it works · Pricing · Security
Resources: Blog · API docs · Help center · Changelog · GitHub
Solutions: For developers · For automation agencies
Company: About · AI info: company facts · Contact
Legal: Terms of service · Privacy policy · Refund policy · Acceptable use · Cookie policy · DPA · Subprocessors
Elsewhere: Status page · Cotlet on GitHub · Cotlet on X · Cotlet on LinkedIn · Cotlet on Product Hunt · Cotlet on Crunchbase · Cotlet on SaaSHub
Cotlet.dev is built by Bart Krawczyk and Jake Slugocki, the founders of Take Group sp. z o.o., ul. Wincentego Janasa 4, 40-855 Katowice, Poland. KRS 0000871398, NIP 6342984273, VAT EU: PL6342984273. About us
Cotlet is an independent service. It is not affiliated with, endorsed by or sponsored by any of the products listed on this site. Product names are trademarks of their respective owners.