Connect Open Pulse to ChatGPT, Claude and Claude Code
Step-by-step setup for each MCP client, the plan requirements that silently block a correct setup, what the OAuth 2.1 flow is doing behind the consent screen, and the prompts that get useful answers rather than a tool-call loop.
What this page answers
What does this post cover?
Step-by-step setup for each MCP client, the plan requirements that silently block a correct setup, what the OAuth 2.1 flow is doing behind the consent screen, and the prompts that get useful answers rather than a tool-call loop.
Who is it for?
Anyone connecting a workspace to an AI assistant, and engineers who want to know what the consent screen is actually doing
Do I need a paid plan to use MCP connectors?
For ChatGPT, Grok and Claude on the web or desktop, yes: custom connectors are a paid-tier feature on all of them, and the option simply does not appear on free plans with no explanation. Claude Code works from any terminal with the CLI installed.
Why is there no tool to create a listener?
Because a listener is a standing commitment to spend on every run. An agent that misreads an instruction and creates several has made a billing decision rather than a formatting mistake, so creation and deletion stay in the interface where plan limits are visible.
What can an AI agent actually do with my workspace?
Read listeners, signals, roll-ups, runs and the quality report; rate a signal; and queue a run if it holds admin scope. It cannot create or delete listeners, change billing, or reach anything outside the workspace you authorised.
At a glance
- Published
- 2026-09-22
- Written for
- Anyone connecting a workspace to an AI assistant, and engineers who want to know what the consent screen is actually doing
- Category
- solution
- Reading time
- 6 minutes
- Keywords
- mcp server setup, connect chatgpt to mcp, claude custom connector, claude code mcp add, model context protocol client
Overview
Open Pulse is a remote MCP server, which means an AI assistant can read your workspace directly rather than you pasting results into a chat. Ask for this week's high-intent signals and get them, in the tool you already work in.
This is the setup for each client, the requirements that will waste your afternoon if nobody tells you about them, and what the authorisation flow is actually doing.
Before you start: the requirement that blocks people
Every client below supports custom MCP connectors on paid plans only. This is the single most common failure, and it fails invisibly: you follow every step correctly and the option is simply not there, with nothing on screen explaining why.
| Client | What you need |
|---|---|
| Claude | claude.ai on the web, or the Claude desktop app |
| ChatGPT | Plus, Pro, Business, Enterprise or Edu, but not Free |
| Grok | A paid tier; custom connectors are not on free |
| Claude Code | Any terminal with the CLI installed |
You also need your workspace's server URL, which is shown in Open Pulse under the MCP connector settings. It is the same URL for every client.
Claude
- Open Settings → Connectors.
- Choose Add custom connector.
- Name it openpulse and paste the server URL.
- Press Connect, then approve the Open Pulse consent screen.
- The tools appear under the connector in any new chat.
Existing chats will not pick up a newly added connector. Start a new one.
ChatGPT
- Open Settings → Apps → Advanced settings and turn on Developer mode.
- Choose Add custom connector.
- Paste the server URL.
- Complete the sign-in step, then enable the connector for your chats.
Developer mode is the step people miss: without it there is no "add custom connector" option at all, and the settings screen gives no hint that one exists.
Grok
- Open grok.com/connectors.
- Choose New connector, then Custom.
- Paste the server URL and complete authentication.
- Grok reads the tool list and offers the tools in conversation.
Claude Code
One command:
- claude mcp add --transport http openpulse <your-server-url>
Then run /mcp inside a session to complete the sign-in. The tools are then available to the agent in that project.
This is the setup worth having if you want an agent to act on signals rather than report them: pulling the week's signals, cross-referencing them against a local file of target accounts, and writing a summary is a single prompt once the server is connected.
What the consent screen is actually doing
Worth understanding, because "paste a URL and click approve" hides a genuinely well-designed flow, and because it explains why the setup is this short.
MCP authorisation builds on the OAuth 2.1 draft plus four RFCs, and the discovery is entirely automatic:
- The client makes a request with no token. The server answers 401 with a WWW-Authenticate header pointing at its resource metadata.
- The client fetches /.well-known/oauth-protected-resource (RFC 9728), which names the authorisation server.
- It fetches /.well-known/oauth-authorization-server (RFC 8414) for the endpoints.
- If it has no client ID, it registers itself dynamically (RFC 7591). This is why you never create an app or paste a client secret.
- It generates PKCE parameters (required, not optional) and includes a resource parameter (RFC 8707) naming the exact server the token is for.
- You approve in the browser, the code comes back, and the token is exchanged.
The resource parameter is the security-relevant part: it binds the token to this server specifically, so a token issued for one MCP server cannot be replayed against another. Our server validates that every token presented was issued for it, and refuses anything else.
One implementation note that cost us an outage class: our /.well-known/oauth-authorization-server document never fails. When the identity provider is unreachable it is derived from the publishable key, which encodes the issuer. It used to return a 502, and since that route is the only place part of the metadata ships, one blip made the whole server scan as though it had never implemented the convention.
What the tools actually do
Reads first, writes last. The agent gets:
| Tool | What it returns |
|---|---|
| list_workspaces | The workspaces you can access |
| list_listeners | What is being watched |
| get_listener | One listener's configuration |
| search_signals | Signals filtered by listener, class, source, score or window |
| get_rollup | One row per competitor, account or place |
| list_runs | Recent runs and their status |
| get_quality_report | Per-query and per-source yield |
| rate_signal | Thumbs up or down, which calibrates the next pass |
| trigger_run | Queues a run now, admin only |
There is no create-listener tool, and that is deliberate. A listener is a standing commitment to spend money on every run. An agent that misreads a sentence and creates six of them has not made a formatting error, it has made a billing decision. Creating and deleting listeners stays in the interface where somebody can see the plan limits.
The prompts that actually work
The most common disappointment is a vague question producing a tool-call loop and a mediocre summary. Two rules fix most of it.
Prefer roll-ups to searches for sales questions. Anything shaped like "who should we contact" or "which competitor is vulnerable" wants get_rollup, which returns the row a salesperson acts on. search_signals returns the evidence underneath it, which is what you want after you have picked a row.
Constrain the window and the class. "Show me signals" is an unbounded request. "People actively leaving a named competitor in the last fortnight" maps cleanly onto a classification and a window and returns something useful.
Prompts that work well in practice:
- "Which of our competitors lost the most users this month, and what were the top three complaint themes for each?"
- "Show me high-intent signals from the last seven days, with the reasoning, and group them by source."
- "Has the quality report got any queries that have never produced a kept signal? List them so I can prune them."
- "Rate the signals I mark as good, then tell me what they have in common."
That last one is worth trying. Ratings feed the next pass, so an agent that helps you rate consistently is improving the filter, not just reading it.
Security boundaries, stated plainly
- The tools are read-only apart from rating a signal and queueing a run.
- An agent can never create or delete a listener, or change billing.
- Tokens are bound to this server by audience validation and the resource parameter, and tokens issued for anything else are rejected.
- Open Pulse asks for no credentials to anything you own. It reads public sources and sends results outward. It does not connect to your CRM or your inbox.
The broader argument for reading signals through an agent rather than a dashboard is in give your AI agent a research tool, and the rest of the machine-readable surface is described in how we made a marketing site that agents can read.
Frequently asked questions
Do I need a paid plan to use MCP connectors?
For ChatGPT, Grok and Claude on the web or desktop, yes: custom connectors are a paid-tier feature on all of them, and the option simply does not appear on free plans with no explanation. Claude Code works from any terminal with the CLI installed.
Why is there no tool to create a listener?
Because a listener is a standing commitment to spend on every run. An agent that misreads an instruction and creates several has made a billing decision rather than a formatting mistake, so creation and deletion stay in the interface where plan limits are visible.
What can an AI agent actually do with my workspace?
Read listeners, signals, roll-ups, runs and the quality report; rate a signal; and queue a run if it holds admin scope. It cannot create or delete listeners, change billing, or reach anything outside the workspace you authorised.
Is my data sent to the AI provider?
Whatever the assistant reads through the tools enters that assistant's context, the same as anything you paste into a chat. The connection is authorised per workspace and scoped to read operations, but the results are visible to the client you connected.
Which prompt style gets the best results?
Ask for roll-ups rather than raw signal searches for anything shaped like a sales question, and always bound the request by time window and classification. Unbounded requests produce tool-call loops; "switching signals from the last fortnight" produces an answer.
Next
- Start the free trial — Creates a workspace. A card is taken up front and nothing is charged during the trial.
- Pricing — What it costs after the trial.
- All posts — The comparisons and the guides, grouped.