Deepline MCP server
115 operations0 webhooks
Find GTM tools, run and manage enrichment Plays and their triggers, deploy Monitors, manage provider credentials, secrets, notifications and billing through Deepline.
Try Deepline with your agent
Connect Deepline, then paste this into Claude, ChatGPT or Cursor. For ideas from other apps, see the use cases.
Using Corsair, show me what you can do in Deepline on my account, then run one read-only example and show me the result.
Deepline tools and triggers
Tools are what your agent can do in Deepline through Corsair's MCP server. Triggers are the Deepline events a routine can start from.
Monitors
- AccessRead-only
Check whether the workspace can use Deepline Monitors
monitors.access
- AvailableRead-only
List deployable monitor tools, or describe the payload, outputs and pricing of one tool
monitors.available
- Batch.GetRead-only
Read a monitor batch operation, such as the plan from a dry-run start or stop
monitors.batch.get
- CheckRead-only
Validate a monitor definition and show its plan and pricing without deploying or spending credits
monitors.check
- DeleteDestructive
Delete a deployed monitor and deprovision its upstream resource; dryRun shows the plan
monitors.delete
- DependentsRead-only
List the published Plays that react to one monitor
monitors.dependents
- DeployDestructive
Deploy a monitor from a definition; creates an upstream provider resource and may spend credits
monitors.deploy
- Fleets.ListRead-only
List Monitor Fleets, or their overview with status counts
monitors.fleets.list
- GetRead-only
Get one deployed monitor by its public key
monitors.get
- HealthRead-only
Read monitor delivery and convergence health
monitors.health
- Jobs.CancelDestructive
Cancel a running Monitor Job
monitors.jobs.cancel
- Jobs.GetRead-only
Get the state, phase and counts of a Monitor Job
monitors.jobs.get
- Jobs.LogsRead-only
Read the logs of a Monitor Job
monitors.jobs.logs
- Ledger.FindingsRead-only
List the recent findings of one monitor
monitors.ledger.findings
- Ledger.GetRead-only
Get the ledger entry of one monitor
monitors.ledger.get
- Ledger.ListRead-only
List monitors with status, findings counts, signals and the Plays each one triggers
monitors.ledger.list
- ListRead-only
List deployed monitors by lifecycle state, with cursor paging
monitors.list
- PreviewDestructive
Sample recent matches for a monitor definition before deploying; each returned match costs credits
monitors.preview
- ReactivateDestructive
Reactivate a disabled monitor, creating a replacement upstream resource; dryRun shows the cost
monitors.reactivate
- StartDestructive
Apply the retained definition of a stopped monitor upstream; may spend credits
monitors.start
- StopDestructive
Stop a monitor upstream while keeping its definition for monitors.start
monitors.stop
- TestRead-only
Validate an explicit event payload against the webhook contract of a deployed monitor; writes, charges and dispatches nothing
monitors.test
- UpdateDestructive
Update the name or payload of a deployed monitor; payload changes can replace the upstream resource and spend credits
monitors.update
Plays
- ActivityRead-only
Count Play runs and delivered Monitor findings by day
plays.activity
- Clear HistoryDestructive
Permanently delete a Play's run history and saved sheet data, keeping its code and versions
plays.clearHistory
- CostRead-only
Estimate credits per row or run for one saved or prebuilt Play (does not start a run)
plays.cost
- DeleteDestructive
Move a workspace Play to Trash and disable its cron and webhook triggers (restorable)
plays.delete
- GetRead-only
Describe one callable Play
plays.get
- ListRead-only
List and search callable Plays
plays.list
- PathsRead-only
List the folder locations of workspace Plays, one page
plays.paths
- PinWrite
Pin or unpin a workspace Play so it sorts first
plays.pin
- RestoreWrite
Restore a workspace Play from Trash
plays.restore
- RunDestructive
Start a saved or prebuilt Play run (may spend credits or perform external actions)
plays.run
- Settings.GetRead-only
Read a workspace Play's location and display name
plays.settings.get
- Settings.UpdateWrite
Change a workspace Play's location or display name; omitted fields stay as they are
plays.settings.update
- Share.GetRead-only
Read a Play's public share page status and revision picker
plays.share.get
- Share.PublishDestructive
Publish or repoint a Play's public share page to a revision (publicly viewable)
plays.share.publish
- Share.RegenerateDestructive
Regenerate a Play's public share page copy with an LLM (rewrites the public page)
plays.share.regenerate
- Share.UnpublishDestructive
Delete a Play's public share page and its cards
plays.share.unpublish
- Share.UpdateWrite
Change share page settings: search indexing, cost and latency display
plays.share.update
- Triggers.GetRead-only
Read a workspace Play's cron schedule and inbound webhook trigger, with the webhook URL
plays.triggers.get
- Triggers.HistoryRead-only
List one trigger's recent runs
plays.triggers.history
- Triggers.InventoryRead-only
List cron and webhook triggers across the workspace, one page
plays.triggers.inventory
- Triggers.UpdateDestructive
Set or remove a Play cron schedule or inbound webhook (optional HMAC or Standard Webhooks signing) and publish it live (scheduled runs may spend credits)
plays.triggers.update
- Versions.ListRead-only
List a Play's saved versions, newest first
plays.versions.list
- Versions.PublishDestructive
Make a Play version live; its cron and webhook triggers then run that version (may spend credits)
plays.versions.publish
Billing
- Auto Recharge.GetRead-only
Read the automatic credit recharge settings
billing.autoRecharge.get
- Auto Recharge.UpdateDestructive
Turn automatic recharge on or off or change its limits (lets Deepline charge the saved card automatically)
billing.autoRecharge.update
- BalanceRead-only
Get workspace credit balance
billing.balance
- CatalogRead-only
Read the published plan catalog, meters and the plan the workspace is on
billing.catalog
- Credit Purchases.CreateDestructive
Buy credits through a payment-gated purchase (spends money)
billing.creditPurchases.create
- Credit Purchases.RecoverWrite
Check or retire an uncertain credit purchase by its original key (never charges)
billing.creditPurchases.recover
- EntitlementsRead-only
Read current workspace provider benefits
billing.entitlements
- Included ComputeRead-only
Read Play compute pricing, included minutes and minutes used this period
billing.includedCompute
- InvoicesRead-only
List subscription invoices and credit purchase receipts, newest first
billing.invoices
- LedgerRead-only
List or export the credit ledger
billing.ledger
- Plan Transitions.CreateDestructive
Start, change, cancel or restore the billing plan (changes what the workspace pays)
billing.planTransitions.create
- PlansRead-only
List plans that can be started or changed to, and check a promo code without redeeming it
billing.plans
- Portal Sessions.CreateWrite
Create a Stripe billing portal link for payment details and invoices (the link grants billing access)
billing.portalSessions.create
- QuoteRead-only
Estimate the maximum credits for Play sandbox sessions (quote only, spends nothing)
billing.quote
- StatusRead-only
Read plan, payment, recharge and invoice state
billing.status
- Subscription.CancelDestructive
Cancel the subscription at period end, or undo a pending cancellation (changes the subscription)
billing.subscription.cancel
- Subscription.StatusRead-only
Read subscription state, renewal and cancellation facts, and remaining credit pools
billing.subscription.status
- Top UpDestructive
Charge the saved payment method and add credits (spends money)
billing.topUp
- UsageRead-only
Get usage and recent charges
billing.usage
- Usage EventsRead-only
Page raw per-request usage events, or look up one by request_id (not a final charge)
billing.usageEvents
Runs
- BillingRead-only
Read credits billed to up to 100 runs and whether they are settling
runs.billing
- ChildrenRead-only
List one page of async child Runs
runs.children
- EventsRead-only
Search structured run events across the workspace or one run
runs.events
- GetRead-only
Get one Play run status
runs.get
- HistoryRead-only
Search and page workspace run history
runs.history
- InputRead-only
Read the exact original input retained for a run
runs.input
- ListRead-only
List Play runs filtered by Play, status or run id
runs.list
- LogsRead-only
Read retained run logs; continue with lastSeq as afterSeq
runs.logs
- ReceiptsRead-only
Read retained tool results for a run by receipt key, without executing tools
runs.receipts
- RerunDestructive
Start a new run from a past run's input and revision (may spend credits)
runs.rerun
- ResumeDestructive
Continue a failed or cancelled run in place (may spend credits)
runs.resume
- StatusRead-only
Read the lifecycle status and progress counters of one run
runs.status
- StopDestructive
Stop one running Play run
runs.stop
- Stop AllDestructive
Stop every active run in the workspace; dryRun only lists them
runs.stopAll
Notifications
- ChannelsRead-only
List Slack channels visible to the connected Deepline Slack app
notifications.channels
- Configuration.ChangeWrite
Save or delete one notification group atomically against the expected workspace version
notifications.configuration.change
- Configuration.GetRead-only
Read notification groups, destinations, rules and the workspace version
notifications.configuration.get
- CreateWrite
Create a named Slack notification for chosen Play events
notifications.create
- DeleteDestructive
Archive a named notification without touching its Slack integration
notifications.delete
- Dlq.GetRead-only
Get one dead-lettered notification delivery
notifications.dlq.get
- Dlq.ListRead-only
List notification deliveries that exhausted their retries
notifications.dlq.list
- Dlq.UpdateDestructive
Retry (send again) or archive one dead-lettered notification delivery
notifications.dlq.update
- ListRead-only
List named notifications, the event catalog, destinations and dead-letter count
notifications.list
- TestDestructive
Send a test message to the Slack target of a notification
notifications.test
- UpdateWrite
Pause or resume a notification, or change its name, events and Slack target
notifications.update
Tools
- Describe ModelRead-only
Describe an AI Gateway model: metadata, provider options and an example input
tools.describeModel
- ExecuteDestructive
Execute one provider-backed tool (may spend credits or perform external actions)
tools.execute
- GetRead-only
Describe one provider-backed tool
tools.get
- ListRead-only
List callable provider tools
tools.list
- ProvidersRead-only
List discoverable providers with tool counts and categories, or only recent additions
tools.providers
- QuoteRead-only
Price an ai_inference, deeplineagent or ai_evaluate payload without running it
tools.quote
- SearchRead-only
Search callable provider tools
tools.search
Integrations
- BalanceRead-only
Read the live balance of the workspace's own provider credential (LeadMagic)
integrations.balance
- Connect LinkRead-only
Get the dashboard link where a workspace admin connects a provider
integrations.connectLink
- Set CredentialsDestructive
Save provider credential fields for the workspace (replaces the stored credential; values never returned)
integrations.setCredentials
- Start OAuthWrite
Start provider OAuth and return the URL a person opens to grant access
integrations.startOAuth
Secrets
- InspectRead-only
Check whether a secret is active for the org or a Play, without its value
secrets.inspect
- ListRead-only
List Play secret metadata (never values)
secrets.list
- SetDestructive
Create or rotate a Play or org secret (replaces the stored value; value never returned)
secrets.set
- Update NoteWrite
Add, replace or clear the non-sensitive note on one secret
secrets.updateNote
Datasets
- HistoryRead-only
Read immutable run row history with coverage and continuation cursor
datasets.history
- RowsRead-only
Read a page of saved dataset rows, optionally scoped to a run
datasets.rows
Workspace
- Members.ListRead-only
List one page of workspace members and their grants
workspace.members.list
- Service Accounts.ListRead-only
List workspace service accounts and key metadata (prefix and last 4 only)
workspace.serviceAccounts.list
Auth
- StatusRead-only
Show the API key, user, workspace and rate-limit tier this plugin is using
auth.status
Db
- QueryDestructive
Run SQL on the workspace customer database; writes can change storage tables
db.query
Executions
- Get By KeyRead-only
Recover a tool execution by its idempotency key
executions.getByKey
Health
- GetRead-only
Check Runtime API availability
health.get
Search
- QueryRead-only
Search tools, monitor types, plays, and docs
search.query
Connect Deepline to Claude, ChatGPT or Cursor
- 1
Connect Deepline in Corsair
Add your Deepline API key once in Corsair. Corsair stores the credential encrypted and makes the Deepline API calls for your agent, so it never goes into a prompt or a config file. Connect Deepline
If your link lets the agent add apps, you can skip this: it gives you a sign-in link the first time it needs Deepline. On a ChatGPT personal plan, connect Deepline here first.
- 2
Add Corsair's MCP server to your agent
Every client uses the same address:
https://mcp.corsair.dev/mcpOpen Customize → Connectors → Add custom connector, paste the address, then select Connect.Add a custom MCP server with the address and choose OAuth. Personal plans get read tools only.Add to CursorOr add this to~/.cursor/mcp.json. Cursor opens Corsair to sign in the first time it connects.{ "mcpServers": { "corsair": { "url": "https://mcp.corsair.dev/mcp" } } }Run this, then run/mcpin Claude Code to sign in.claude mcp add --transport http corsair https://mcp.corsair.dev/mcpCodex, Gemini CLI, VS Code, Zed and other clients with OAuth sign-in work the same way. Clients without it, such as GitHub Copilot's coding agent, use a link's own address and an API key. Setup for every client
- 3
Sign in and choose what needs your approval
Your client opens Corsair to sign in. Pick a link: the apps your agent can use and a mode that decides what it may do without asking. In the default Cautious mode, Deepline's read and write operations run right away and its 32 destructive operations wait for your approval. Strict mode asks before every write and refuses destructive operations, and Read mode allows reads only.
FAQ
Install corsair and @corsair-dev/deepline, add the Deepline plugin, and connect with an API key. Then call methods like corsair.deepline.api.auth.status() from your app. Corsair keeps each tenant isolated.
Yes. Deepline requires an API key from your Deepline account or developer dashboard. Create credentials with access that matches what you automate, then paste them into Corsair or connect through managed auth in Hub.
Yes. One Corsair client can hit Deepline, GitHub, Notion, Stripe, and the rest. Trigger Deepline from a webhook, push updates to Slack, or chain actions across plugins. Same tenant context across integrations, plain TypeScript.
Teams use Corsair for find GTM tools, run and manage enrichment Plays and their triggers, deploy Monitors, manage provider credentials, secrets, notifications and billing through Deepline. Common builds include sync jobs, admin tooling, and agents that call Deepline APIs from your backend.
The SDK is open source. Self-host and you pay Corsair nothing for the integration layer. Hub bills for hosted auth and execution, not per Deepline API call. Hub has a generous free tier. For hobby projects you can probably run the whole thing on free. Paid plans start to matter with more tenants or heavier traffic.
We maintain @corsair-dev/deepline in the open repo. When Deepline ships API changes, upgrade the package. Your call sites stay typed. You should not need to rewrite the integration layer each time.
No. Corsair does not read your Deepline data or hold individual access credentials. Credentials stay in your database. API calls run on your instance. We do not store Deepline data on your behalf.
Yes. Add https://mcp.corsair.dev/mcp as a custom connector or MCP server in Claude, ChatGPT, Cursor, VS Code, Claude Code, Codex or any client that supports remote MCP servers with OAuth sign-in, then sign in to Corsair. On ChatGPT personal plans, custom servers currently get read tools only.
Yes. Put Deepline on a link in Strict mode and every write waits for your approval, while destructive actions are refused. The default Cautious mode asks only before the 32 destructive Deepline operations. You can change a link's mode at hub.corsair.dev/me.