Merge Unified's Pylon integration Merge Agent Handler's Pylon MCP server

Connect your AI Agents to Pylon in minutes

Merge lets you securely connect your agents to Pylon and thousands of tools instantly

Start building for free

Read docs

Pylon

Learn more about Merge Agent Handler

Available tools

list_accounts

List Pylon accounts with optional pagination.

get_account

Get a specific Pylon account by ID with full details.

create_account

Create a new Pylon account with name, domain, and optional tags. Use list_tags to find valid tag values before assigning them.

update_account

Update an existing Pylon account's name, domain, or tags. Use searchaccounts to find accountid and list_tags for valid tag values.

delete_account

Delete a Pylon account by ID.

search_accounts

Search Pylon accounts with filters for query, domain, or tags.

list_contacts

List Pylon contacts with optional pagination.

get_contact

Get a specific Pylon contact by ID with full details.

create_contact

Create a new Pylon contact with name, email, phone, and optional account association. Use searchaccounts or listaccounts to find a valid account_id before associating.

update_contact

Update an existing Pylon contact's name, email, phone, or account association. Use searchcontacts to find contactid and searchaccounts for a valid accountid.

delete_contact

Delete a Pylon contact by ID.

search_contacts

Search Pylon contacts with filters for query, email, account, or tags.

list_issues

List Pylon issues within a time range. Requires starttime and endtime in ISO 8601 format.

get_issue

Get a specific Pylon issue by ID with full details.

create_issue

Create a new Pylon issue. Requires title and bodyhtml. Without destinationmetadata the issue is internal and replytoissue will reject it. For a customer-facing issue pass destinationmetadata: 'customerportal' needs nothing else, 'email' also needs a sender address and an accountid. Use searchaccounts for accountid, searchusers for assigneeid, listteams for team_id.

update_issue

Update an existing Pylon issue's title, body, state, priority, assignee, team, or tags. Use searchissues or listissues to find issueid, searchusers or listusers for assigneeid, listteams for teamid, and list_tags for valid tag values.

delete_issue

Delete a Pylon issue by ID.

search_issues

Search Pylon issues with filters for state, priority, account, assignee, team, or tags.

reply_to_issue

Reply to a Pylon issue. Sends a customer-visible response, so it only works on issues with a customer channel: an internal issue is rejected with 'You cannot reply to an internal issue' -- use addissuenote there instead. Use searchissues for issueid and getissuemessages for message_id; both required.

create_issue_thread

Create a thread on a Pylon issue for internal notes. Returns threadid for use with addissue_note.

add_issue_note

Add an internal note to a Pylon issue. Notes are not visible to customers. Use searchissues to find issueid, and createissuethread first if the note belongs in a specific thread.

get_issue_messages

Get all messages for a Pylon issue, including replies and internal notes.

list_knowledge_bases

List Pylon knowledge bases available to this account, with optional pagination (cursor only; this endpoint has no limit param).

list_articles

List articles within a Pylon knowledge base with optional pagination. Use listknowledgebases to find knowledgebaseid.

get_article

Get a specific Pylon knowledge base article by ID with full details. Use listknowledgebases for knowledgebaseid and listarticles for articleid.

update_article_draft

Update the draft content of a Pylon knowledge base article (title, bodyhtml, slug, collection, visibilityconfig). Updates the default-language article only. Never affects the live published version. Use listknowledgebases for knowledgebaseid and listarticles or getarticle for article_id. Requires at least one content field.

update_article_published

Update a Pylon knowledge base article and push the change to its live published version (title, bodyhtml, slug, collection, visibilityconfig). Updates the default-language article only. Use listknowledgebases for knowledgebaseid and listarticles or getarticle for article_id. Requires at least one content field.

create_article

Create a Pylon knowledge base article. Requires title, bodyhtml and authoruserid. Created as a draft unless ispublished is true. Use listknowledgebases for knowledgebaseid and listusers for authoruser_id.

delete_article

Delete a Pylon knowledge base article. Use listknowledgebases for knowledgebaseid and listarticles for articleid.

list_tags

List all Pylon tags. This endpoint is not paginated: it returns every tag, so limit only trims the response. Use the returned values for the tags parameter on searchissues and searchaccounts.

get_tag

Get a specific Pylon tag by ID with full details.

create_tag

Create a new Pylon tag with value, objecttype, and optional hexcolor.

delete_tag

Delete a Pylon tag by ID.

list_teams

List all Pylon teams. This endpoint is not paginated: it returns every team, so limit only trims the response.

get_team

Get a specific Pylon team by ID with full details including members.

list_users

List all Pylon users (internal team members). This endpoint is not paginated: it returns every user, so limit only trims the response. Set includedeactivated to include deactivated users. Use searchusers for a server-side filtered, bounded result.

get_user

Get a specific Pylon user by ID with full details.

search_users

Search Pylon users with filters for query, email, role, team, or active status.

get_current_user

Get the currently authenticated Pylon user. Returns the user the API token belongs to; the owning organization is exposed as organizationid and organizationname.

validate_credential

Validate Pylon credentials. Verifies credentials during setup.

View all tools by creating a free account See more tools

How to set up Merge Agent Handler

In an mcp.json file, add the configuration below, and restart Cursor.

Learn more in the official documentation ↗

1{
2  "mcpServers": {
3    "agent-handler": {
4      "url": "https://ah-api-develop.merge.dev/api/v1/tool-packs/{TOOL_PACK_ID}/registered-users/{REGISTERED_USER_ID}/mcp",
5      "headers": {
6        "Authorization": "Bearer yMt*****"
7      }
8    }
9  }
10}
11

Copy Code

Open your Claude Desktop configuration file and add the server configuration below. You'll also need to restart the application for the changes to take effect.

Make sure Claude is using the Node v20+.

Learn more in the official documentation ↗

1{
2  "mcpServers": {
3    "agent-handler": {
4      "command": "npx",
5      "args": [
6        "-y",
7        "mcp-remote@latest",
8        "https://ah-api-develop.merge.dev/api/v1/tool-packs/{TOOL_PACK_ID}/registered-users/{REGISTERED_USER_ID}/mcp",
9        "--header",
10        "Authorization: Bearer ${AUTH_TOKEN}"
11      ],
12      "env": {
13        "AUTH_TOKEN": "yMt*****"
14      }
15    }
16  }
17}

Copy Code

Open your Windsurf MCP configuration file and add the server configuration below.
Click on the refresh button in the top right of the Manage MCP server page or in the top right of the chat box in the box icon.

Learn more in the official documentation ↗

1{
2    "mcpServers": {
3      "agent-handler": {
4        "command": "npx",
5        "args": [
6          "-y",
7          "mcp-remote@latest",
8          "https://ah-api.merge.dev/api/v1/tool-packs/<tool-pack-id>/registered-users/<registered-user-id>/mcp",
9          "--header",
10          "Authorization: Bearer ${AUTH_TOKEN}"
11        ],
12        "env": {
13          "AUTH_TOKEN": "<ah-production-access-key>"
14        }
15      }
16    }
17  }

Copy Code

In Command Palette (Cmd+Shift+P on macOS, Ctrl+Shift+P on Windows), run "MCP: Open User Configuration".

You can then add the configuration below and press "start" right under servers. Enter the auth token when prompted.

Learn more in the official documentation ↗

1{
2  "inputs": [
3    {
4      "type": "promptString",
5      "id": "agent-handler-auth",
6      "description": "Agent Handler AUTH_TOKEN", // "yMt*****" when prompt
7      "password": true
8    }
9  ],
10  "servers": {
11    "agent-handler": {
12      "type": "stdio",
13      "command": "npx",
14      "args": [
15        "-y",
16        "mcp-remote@latest",
17        "https://ah-api-develop.merge.dev/api/v1/tool-packs/{TOOL_PACK_ID}/registered-users/{REGISTERED_USER_ID}/mcp",
18        "--header",
19        "Authorization: Bearer ${input:agent-handler-auth}"
20      ]
21    }
22  }
23}

Copy Code

FAQs on using Merge's Pylon MCP server

What is a Pylon MCP server?

It’s an MCP server that lets you access data and functionality from a Pylon instance via tools. Your agents can call these tools to create accounts, update contacts, fetch issues, and more.

Pylon offers their own “official” MCP server. You can also access additional MCP servers from 3rd-party platforms, like Merge Agent Handler.

What are some common ways to use a Pylon MCP server?

Here are a few popular use cases:

Support ticket triage and routing: Pull ticket context, customer metadata, and history into an agent, then auto-classify issue type, priority, integration, and likely owner

Draft effective replies: Generate a suggested response grounded in the ticket thread and account details, including clarifying questions and recommended next steps

Runbook-driven troubleshooting: Follow a standardized diagnostic playbook to gather required info, run checks, and decide next actions so resolution is more consistent and faster

Ticket pattern analysis: Analyze historical tickets to identify top drivers and emerging themes, then output a ranked list with frequency and representative examples

Knowledge base gap detection and help center improvements: Detect missing/outdated/duplicate documentation behind repeat tickets and convert those signals into a prioritized docs backlog

What are the most impactful MCP tools in a Pylon MCP server?

Here are some worth calling:

1. <code class="blog_inline-code">List_accounts</code>: Returns a paginated list of customer accounts with key metadata (name, ID, status, owner), so the model can browse and reference accounts in context.

2. <code class="blog_inline-code">Get_account</code>: Fetches detailed information for a single account by ID (contacts, MRR, lifecycle stage, notes), enabling precise, account‑specific reasoning.

3. <code class="blog_inline-code">List_contacts</code>: Lists contacts across accounts (with filters like account_id, role, status), so the model can identify the right people associated with a customer.

4. <code class="blog_inline-code">Get_contact</code>: Retrieves a single contact’s full profile (email, role, account, last activity), supporting personalized responses and outreach suggestions.

5. <code class="blog_inline-code">List_tickets</code>: Returns support tickets (with filters like account_id, status, priority), allowing the model to see open issues and historical context for a customer.

6. <code class="blog_inline-code">Get_ticket</code>: Fetches the full detail of a specific ticket (description, logs, comments, SLA), so the model can troubleshoot or summarize the case accurately.

7. <code class="blog_inline-code">List_opportunities</code>: Lists sales opportunities (with filters for stage, owner, account_id), enabling pipeline analysis and deal‑specific recommendations.

8. <code class="blog_inline-code">Create_note</code>: Creates an internal note linked to an account/contact/ticket, so the model can log summaries, next steps, or call notes back into your system.

What makes Merge Agent Handler’s Pylon MCP server better than alternative Pylon MCP servers?

Merge Agent Handler provides platform-level advantages:

End-to-end integration lifecycle management: Merge Agent Handler handles authentication, credential management, rate limiting, and error handling across all connectors, reducing infrastructure overhead for developers

Enterprise-grade security with DLP: A built-in security gateway scans tool inputs and outputs for sensitive data and can block, redact, or mask data based on configurable rules

Real-time observability and auditability: The platform logs every tool call and underlying API request, enabling fully searchable logs for debugging, auditing, and optimization

Evaluation Suite for pre-production validation: Developers can validate tool behavior against expected outcomes and identify performance gaps before deploying to production

Connector Studio for customization: Teams can modify existing connectors or build custom tools while maintaining production-grade reliability

Can I set custom security rules for Pylon tool calls in Merge Agent Handler?

Yes, Merge Agent Handler includes a security gateway with data loss protection (DLP) that allows you to define custom rules for tool inputs and outputs.

Here are some rules you can set:

Prevent sensitive data leaks in customer conversations. Block or redact tool calls that would add PII, credentials, or payment details into Pylon comments, internal notes, or outbound replies

Avoid off-brand or incorrect customer replies. Require approval before any Pylon tool call that sends a customer-facing message or updates a customer-visible ticket field (status, priority, SLA)

Reduce operational risk from “mass edits”. Block or step-up approve Pylon tool calls that apply changes to more than N tickets at once (e.g., bulk tagging, bulk status changes, merges, etc.)

Enforce least-privilege access for the support workflow. Allow the agent to use only the specific Pylon tools needed for the workflow (e.g., “draft reply” and “suggest tags”), and deny admin/config tools by default

Prevent cross-customer mistakes. Deny Pylon tool calls unless the ticket’s workspace/account/customer identifier matches the active end-user/customer context

How can I get started with Merge Agent Handler’s Pylon MCP server?

1. Create (or sign in to) your Merge Agent Handler account and open the dashboard.

2. Create a Tool Pack for your Pylon workflow (this is the bundle of tools your agent will be allowed to call).

3. Add the Pylon connector to that Tool Pack, and start by enabling only the specific tools you need.

4. Set your security and policy baseline for Pylon tool calls (DLP rules, approvals for risky actions, rate limits, etc.).

5. Create a Registered User (the execution identity boundary), then complete the Pylon authentication flow for that Registered User so tool calls run under the right permissions.

6. Copy the MCP endpoint URL for your Tool Pack and Registered User and add it to your MCP client (Cursor, Claude Desktop, VS Code, Windsurf, etc.).

7. Run a simple “first tool call” in a test workflow, then check logs and iterate on Tool Pack scope and rules before rolling into production.