Anthropic's Claude and CustomGPT Integration

Demonstration of how apps are integrated in WEXTL workflow builder[4]

Keep Anthropic's Claude and CustomGPT matched by integrating their events and actions. When ai events happen in Anthropic's Claude, update the CustomGPT automatically - without any line of code in visual workflow builder.

  • 3-day trial
  • Cancel anytime
  • Up to 10 minutes quick set up
  • GDPR
  • Data Encryption
  • 2FA
  • Local Data Region

Learn about Anthropic's Claude and CustomGPT

Detailed info you need to know about Anthropic's Claude and CustomGPT integrations in WEXTL. Explore which modules apps support and how you connect them.

  • Anthropic's Claude

    • AI
    • 21 triggers
    • 60 searches
    • 81 actions

    Anthropic Claude models for reasoning, writing, and analysis.[5]

  • CustomGPT

    • AI
    • 4 triggers
    • 57 searches
    • 71 actions

    AI agents trained on your own content.[6]

How to connect Anthropic's Claude to CustomGPT

Set up your Anthropic's Claude and CustomGPT integration in 3 simple steps[2]. No coding required. Up to 10 minutes setup.

How does the integration between Anthropic's Claude and CustomGPT work in WEXTL?

WEXTL exposes actions and triggers for Anthropic's Claude and CustomGPT that help you connect apps and automate repetitive tasks[1]. A trigger is an event that happens, and an action is a step that performs when the event has occurred.

Anthropic's Claude triggersCustomGPT actions

Anthropic's Claude modules

  • Add a Federation Rule Workspace Enables a federation rule for one more workspace. You get back the Workspace name the rule now covers (admin only).
  • Add a Workspace to a Service Account Gives a service account access to a workspace with the Workspace Role you assign. The response says whether the membership is Implicit (admin only).
  • Archive a Credential Marks a credential kept in a vault as archived and gives you the Archived at time.
  • Archive a Deployment Archives a deployment. List Deployments leaves archived ones out unless Include Archived is on.
  • Archive a Federation Issuer Puts a federation issuer into the archive. The output lists who did it under Archived by actor ID (admin only).
  • Archive a Federation Rule Archives a federation rule. The response lists the Attributes the rule used and when it was archived (admin only).
  • Archive a Memory Store Marks a memory store as archived. Output includes the Archived at time.
  • Archive a Service Account Moves a service account to the archive and returns its Organization role (admin only).
  • Archive a Session Archives an agent session and returns its Status and Outcome evaluations.
  • Archive a Tunnel Sets a tunnel aside as archived and reports the Domain it was serving (admin only).
  • Archive a Tunnel Certificate Archives one certificate on a tunnel. The Fingerprint in the response confirms which one was archived (admin only).
  • Archive a Vault Archives a vault that holds credentials. The response lists its Display name.
  • Archive a Workspace Archives an organization workspace and reports its Data residency along with the Archived at time (admin only).
  • Archive an Agent Sets an agent aside as archived. Lists leave archived agents out until Include Archived is turned on.
  • Archive an Environment Moves an environment definition to the archive and reports its Config.
  • Create a Credential Saves a credential into a vault from your Authentication details. Add a Display Name to recognize it later.
  • Create a Deployment Sets up a deployment that runs an Agent inside an Environment ID. Use Schedule to set when it runs, Initial Events to open each session with user messages and Vault IDs to give the agent stored credentials.
  • Create a Federation Issuer Adds a token issuer you trust, identified by its Issuer URL. Turn on Check JTI to reject replayed tokens, and use Max JWT Lifetime Seconds to cap assertion age (admin only).
  • Create a Federation Rule Builds a rule that trades tokens from an Issuer ID for access with an OAuth Scope. Turn on Applies To All Workspaces or name one Workspace ID, and use Token Lifetime Seconds to set how long minted access tokens last (admin only).
  • Create a Memory Writes a new memory into a store from Content, up to 100 kB of UTF-8 text. Use Path to place it in a hierarchy of paths that starts with a slash.
  • Create a Memory Store Creates a memory store called Name. A Description of up to 1024 characters is shown in the system prompt of agents the store is attached to.
  • Create a Message Sends Messages to a Claude model and returns its reply and Stop reason. Use Max Tokens to stop after a number of tokens, Temperature to set sampling between 0 and 2 and System Prompt to give it instructions.
  • Create a Service Account Adds a service account labelled by Name. Use Description to record its purpose and Organization Role, which defaults to developer, to set its access level (admin only).
  • Create a Service Account Workspace Member Enrolls a service account in a workspace under the Workspace Role of your choice. The output echoes the Service account ID (admin only).
  • Create a Session Creates a session in an Environment ID, optionally running an Agent. Use Resources to mount repositories or files and Vault IDs to hand it stored credentials.
  • Create a Skill Uploads a skill from Files, which must share one top-level directory with a SKILL.md at its root. Add a Display Title to name it.
  • Create a Skill Version Uploads a further version of an existing skill from Files. The new Version number shows in the output.
  • Create a Tunnel Certificate Registers a CA Certificate PEM on a tunnel. It must hold exactly one certificate and no private-key material (admin only).
  • Create a User Profile Adds a user profile under a Name. Use Relationship to say how this entity is related to your platform, External ID for your own identifier and Metadata for key-value data.
  • Create a Vault Creates a vault for holding credentials, using a Display Name of 1 to 255 characters. Use Metadata to attach key-value data.
  • Create a Workspace Adds a workspace to your organization under a Workspace Name. Use Data Residency to set where its data is stored, External Key ID to attach an encryption key configuration and Tags to label it (admin only).
  • Create a Workspace Member Gives a user a Workspace Role within a workspace. The response shows the User ID that was added (admin only).
  • Create an Agent Defines an agent from a Model and a Name. Use System to give it a system prompt and MCP Servers to connect up to 20 MCP servers. Add Skills and Tools to shape what it can do.
  • Create an Enrollment URL Generates an enrollment link for a user profile. The response gives the expiry as Expires at.
  • Create an Environment Adds an environment that defines the container sessions run in. Use Config for its setup and Scope for who can see it.
  • Create an External Key Registers a customer-managed encryption key under a Display Name. Describe how your key provider is reached with Provider Config (admin only).
  • Create an Invite Invites a person by Email into your organization with the Role they should hold. The output reports the invitation Status and its Expires at time (admin only).
  • Delete a Credential Removes one stored credential from the vault that holds it.
  • Delete a File Removes an uploaded file from your Claude account.
  • Delete a Memory Deletes a single memory entry from a memory store.
  • Delete a Memory Store Deletes an entire store of agent memories.
  • Delete a Service Account Workspace Member Removes a service account from one workspace of your organization (admin only).
  • Delete a Session Removes one agent session from your Anthropic workspace.
  • Delete a Skill Removes a skill you published earlier from your workspace.
  • Delete a Skill Version Deletes a specific uploaded revision of a skill.
  • Delete a User Revokes the membership of a person in the organization (admin only).
  • Delete a Vault Deletes a vault that stores credentials for agents.
  • Delete a Workspace Member Removes a user from the membership list of one workspace (admin only).
  • Delete an Environment Takes an environment definition, with its container configuration, out of your workspace.
  • Delete an External Key Deletes a registered customer-managed encryption key from your organization (admin only).

CustomGPT modules

  • Activate a Persona Version Switches an agent to one of its saved persona versions so that version becomes the active persona. Output includes Updated.
  • Add a Source Adds content for an agent to learn from, through a Source Type of Website or Sitemap, Website with Images or Files. Use Text Recognition to read text from images and scans, and Anonymize Content to hide personal data in uploads.
  • Add an External API Connects an outside API to an agent from its OpenAPI schema, supplied as an OpenAPI Schema URL or an OpenAPI Schema File but not both. A Name and the API Key for that service complete it.
  • Assign Child Agents Makes other agents the children of a master agent. Attach one or several agents through Child Agents.
  • Assign Users to Tags Grants chosen team members access to the tags of an agent. Use User Emails to name the members for each tag.
  • Cancel an Invitation Cancels an invitation that was sent to join your CustomGPT team.
  • Create a Conversation Starts a conversation with an agent. Give it a Name, or leave that empty and the agent names it. Output includes the Session ID used to send messages.
  • Create a Custom Action Connects an MCP server to an agent as an action, from a Name and the MCP URL. A Description or Token is available if the server needs them. Turn on Require Tool Confirmation to ask before a tool runs.
  • Create a License Issues a license for an agent from a Name. Output includes the License Key.
  • Create a Role Adds a team role with a Name that cannot match a built-in role. Use Scope to limit it to Specific Agents or All Agents, and Permissions to say what its members may do.
  • Create a Task Starts a background task for an agent from a first Prompt. Choose a Model and Capability, add Files, and read the answer later with Get Task.
  • Create an Agent Builds a new CustomGPT agent from a Name. Use Website, Sitemap, YouTube or Vimeo Link for web content and Files for uploads, turn on Master Agent to route questions to child agents, and turn on Plan and Act so it plans first.
  • Create Tags Makes tags from a list of Names, each up to 20 characters and not starting with a digit. Output lists the ID and Name of each tag.
  • Delete a Conversation Removes one conversation held with an agent.
  • Delete a Custom Action Removes a custom MCP action from an agent.
  • Delete a Document Removes a document from the indexed pages of an agent.
  • Delete a License Removes a license found by its numeric License ID.
  • Delete a Role Deletes one role from your CustomGPT team.
  • Delete a Source Deletes a content source from an agent.
  • Delete a Sub-Persona Override Deletes the override saved for a sub-persona on one persona version. Output includes Is Generating, Generated and Agent Overrides.
  • Delete a Tag Removes a tag that was created for an agent.
  • Delete a Task Deletes a task that an agent ran in the background.
  • Delete a Task File Deletes a file attached to a message inside a task.
  • Delete an Agent Deletes an agent in your CustomGPT account.
  • Delete an External API Deletes an external API connection from an agent.
  • Delete Webpage Context Pages Removes webpage context pages. Turn on Delete All Pages to remove every page of the agent.
  • Download a Task File Downloads a file that belongs to a message of a task. Output includes the File and its Content Type.
  • Download an Invoice Downloads a CustomGPT billing invoice as a file. Output includes the File, File Name, Size (Bytes) and Content Type.
  • Export a Conversation Saves a conversation to a file. Choose Format as CSV, JSON, Excel or PDF, and narrow what is included with Start Date, End Date and Maximum Rows.
  • Export a Task Saves a task to a file. Pick CSV, JSON, Excel or PDF for Format.
  • Export Leads for a Conversation Writes the leads from one conversation to a file. Set Mode to Every Captured Lead or One Merged Lead, and Format to CSV, JSON, Excel or PDF.
  • Get a Billing Plan Shows the CustomGPT plan on your account along with its Name and Status. Output also gives the billing Interval and the dates the current period runs between.
  • Get a Citation Source Preview Loads a preview of the source behind a citation, using the Citation Source ID found in the citations of a message.
  • Get a Collated Lead Merges the leads captured in one conversation into a single record with Name, Email, Phone Number and Company.
  • Get a License Reads a license by its numeric License ID, returning Name, Key and Created At.
  • Get a Message Reads a message from a conversation, with the User Query, OpenAI Response and Citations. Turn on Include Insights to add customer intelligence data where your plan has it.
  • Get a Message Trust Score Rates how far an answer can be trusted, giving an Overall Status and a Stakeholder Analysis.
  • Get a Persona Version Reads one persona version with its Instructions, Is Active flag and Created At.
  • Get a Skill Version Reads a version of a skill, found by Skill ID and Version, returning Instructions and Is Active. CustomGPT has no list of skills, so copy the Skill ID from the dashboard.
  • Get a Tag Reads one tag, giving its ID, Name and Users.
  • Get a Task Reads a task, showing its Status, Events and Result. Run it after a task starts to read the answer.
  • Get a Task Message Reads a single message that belongs to a background task.
  • Get Account Limits Compares your CustomGPT plan limits with what you use now. Output includes Max Projects Num, Current Projects Num, Max Queries and Current Queries.
  • Get Agent Settings Fetches the chat settings of an agent, such as Chatbot Avatar, Default Prompt, Example Questions and Response Source.
  • Get Agent Statistics Reports how far an agent has crawled and indexed its content. Output includes Pages Found, Pages Crawled, Total Queries and Total Storage Credits Used.
  • Get an Agent Reads one agent, including its Agent Name, Type, Is Master flag and Shareable Link.
  • Get an External API Reads one external API connection of an agent: Host, Base URL, Auth Type, Is Enabled and Last Used At.
  • Get Analytics Chart Data Returns chart figures for an agent, for one or more Charts. Set Interval to Weekly or Daily to change how the numbers are grouped over time.
  • Get Citation Metadata Reads what is shown for a citation: its Title, Description, Summary and Image. Take the Citation ID from the citations of a message.
  • Get Conversation Analytics Reports conversation numbers for an agent. Use Metrics to ask for Total Conversations, Average Queries per Conversation or both.

Explore alternative integrations

There are other apps[9] supported by WEXTL that can be paired with Anthropic's Claude and CustomGPT integrations. Pick any from the list below if the current combination isn't the one you need.

Automate Anthropic's Claude and CustomGPT on WEXTL®

Claim your 3-day trial to build Anthropic's Claude with CustomGPT integration and access every WEXTL® feature.

Cancel anytime.

Frequently asked questions

Common questions about connecting Anthropic's Claude and CustomGPT.

How do I connect Anthropic's Claude and CustomGPT?

Connect your Anthropic's Claude and CustomGPT accounts once, pick a trigger in one app and an action in the other, then map the fields so data flows automatically between them.[3]

What can I automate between Anthropic's Claude and CustomGPT?

Example: when the "Watch Agent Versions" trigger in Anthropic's Claude fires, WEXTL can "Activate a Persona Version" in CustomGPT. Anthropic's Claude has 21 triggers (including Watch Agent Versions, Watch Agents, and Watch Credentials). CustomGPT has 128 actions (including Activate a Persona Version, Add a Source, and Add an External API). Add filters and transforms between steps when you need them.[1]

Can Anthropic's Claude trigger actions in CustomGPT, and CustomGPT in Anthropic's Claude?

Yes. The integration works in both directions. Use the triggers of each app to start a workflow and the actions of the other app to update it.[2]

Do I need to write code to integrate Anthropic's Claude and CustomGPT?

No. Anthropic's Claude-to-CustomGPT automations are built visually on the canvas. You can optionally add custom JavaScript with Functions when you need advanced logic.[4]

Does Anthropic's Claude integrate with CustomGPT in real time?

Yes. Trigger-based automations start the moment Anthropic's Claude (or CustomGPT) fires a new event. Webhook-backed apps run within seconds, and polling-only apps re-check on a schedule you control.[7]

How long does it take to set up a Anthropic's Claude + CustomGPT integration?

A first automation typically takes 5–10 minutes: connect the credential of each app once, drop a trigger + action onto the canvas, map the fields, and turn the workflow on.

Do I need a paid plan to connect Anthropic's Claude with CustomGPT?

You can build your Anthropic's Claude + CustomGPT automation on a 3-day trial with every feature unlocked. Cancel anytime.[8]

References

List of sources used to gather information about the Anthropic's Claude and CustomGPT integration.

  1. Anthropic's Claude and CustomGPT integration source code, module specs and credentials implementation.
  2. Our Anthropic's Claude and CustomGPT integration documentation, structured input and output parameters, scopes, and authentication methods.
  3. Official vendor API documentation for Anthropic's Claude and CustomGPT, used to verify field meanings, authentication flows, and anything our own specs don't fully capture.
  4. WEXTL® visual workflow builder, anonymized data from our technical support team.
  5. Anthropic's Claude integration information
  6. CustomGPT integration information
  7. WEXTL® Scheduling
  8. WEXTL® Pricing and Trial
  9. WEXTL® Apps Library

How we write our apps documentation

Explore how WEXTL® crafted Anthropic's Claude and CustomGPT integration page documentation. Used sources, the role of AI, and how the team validates the result.

Explore Anthropic's Claude and CustomGPT, or get started