Official
Expose 8,000+ Zapier integrations as tools your AI agent can call.
zapier, automation, integrations, workflow
5 minLow
Activepieces
Open-source workflow automation with a built-in MCP server — build, publish, and debug flows from an AI agent.
Quick answer
What it does
Exposes about 52 ap_* tools in seven categories: Discovery (read-only exploration of flows, pieces, connections, tables and runs, plus semantic search over the action and trigger catalog), Flow Management, Flow Building, Router & Branching, Annotations, Tables, and Testing & Runs.
Best for
- Building flows from a natural-language description
- Debugging failed runs
- Read-only automation discovery for agents
- Table CRUD from chat
Not for
- Teams not already running Activepieces
- Setups needing a static API key rather than OAuth
Source: top-mcps.com/mcp/activepieces — verified Sep 2026
Setup recipe
Pick your client, then follow the three steps.
- 1
Install
claude_desktop_config.jsonjson{ "mcpServers": { "activepieces": { "command": "npx", "args": [ "-y", "mcp-remote", "https://cloud.activepieces.com/mcp" ] } } }Claude Desktop launches stdio servers, so mcp-remote bridges to the hosted endpoint and opens the sign-in in your browser. On Pro, Max, Team and Enterprise you can skip the file: Settings → Connectors → Add custom connector → https://cloud.activepieces.com/mcp. Authorization happens in the browser on first use (OAuth); no token goes in the config file.
CLI or .mcp.jsonshellclaude mcp add --transport http activepieces https://cloud.activepieces.com/mcpThen run /mcp inside Claude Code to finish the sign-in. Authorization happens in the browser on first use (OAuth); no token goes in the config file.
.cursor/mcp.jsonjson{ "mcpServers": { "activepieces": { "url": "https://cloud.activepieces.com/mcp" } } }Global path: ~/.cursor/mcp.json. Cursor shows "Needs login" once it connects — click it to authorize. Authorization happens in the browser on first use (OAuth); no token goes in the config file.
.vscode/mcp.jsonjsonc{ "servers": { "activepieces": { "type": "http", "url": "https://cloud.activepieces.com/mcp" } } }VS Code uses the "servers" key. Run MCP: List Servers → Start Server and allow the sign-in. Authorization happens in the browser on first use (OAuth); no token goes in the config file.
~/.codeium/windsurf/mcp_config.jsonjson{ "mcpServers": { "activepieces": { "serverUrl": "https://cloud.activepieces.com/mcp" } } }Windsurf uses "serverUrl" for hosted servers. Open via Cascade → hammer icon → Configure. Authorization happens in the browser on first use (OAuth); no token goes in the config file.
cline_mcp_settings.jsonjson{ "mcpServers": { "activepieces": { "command": "npx", "args": [ "-y", "mcp-remote", "https://cloud.activepieces.com/mcp" ] } } }Bridged through mcp-remote. Open via the Cline sidebar → MCP Servers → Edit. Authorization happens in the browser on first use (OAuth); no token goes in the config file.
~/.continue/config.jsonjson{ "experimental": { "modelContextProtocolServers": [ { "transport": { "type": "streamable-http", "url": "https://cloud.activepieces.com/mcp" } } ] } }Continue takes the hosted URL directly in the transport block. Authorization happens in the browser on first use (OAuth); no token goes in the config file.
CLIshellcodex mcp add activepieces --url https://cloud.activepieces.com/mcpCodex detects OAuth support and opens the browser to authorize. Authorization happens in the browser on first use (OAuth); no token goes in the config file.
~/.config/zed/settings.jsonjsonc{ "context_servers": { "activepieces": { "command": { "path": "npx", "args": [ "-y", "mcp-remote", "https://cloud.activepieces.com/mcp" ] } } } }Bridged through mcp-remote under "context_servers". Settings live-reload on save. Authorization happens in the browser on first use (OAuth); no token goes in the config file.
ChatGPT → Settings → Connectors (remote URL)json{ "name": "Activepieces", "transport": "http", "url": "https://cloud.activepieces.com/mcp" }Paste this into Settings → Connectors → Add custom connector. Requires ChatGPT Pro / Team / Enterprise / Edu with Developer mode enabled.
- 2
Set required secrets
No credentials required — this MCP runs over stdio without authentication.
- 3
Try a minimum working prompt
Minimum working prompt pending verification. Try any prompt from the MCP’s README once installed.
Tools & permissions
| Tool | Description | Args | Side effects |
|---|---|---|---|
ap_list_flows | List flows in the current project with status, trigger type, and published state. | limit?: number, status?: string, name?: string | Read |
ap_flow_structure | Get the full structure of a flow: step tree, configuration status, step input values, router branch conditions, and valid insert locations. | flowId: string | Read |
ap_read_step_code | Read the full source code, package.json, and input mappings of a CODE step. | flowId: string, stepName: string | Read |
ap_read_step_settings | Read the full untruncated settings of any step, including the trigger: piece input, action/trigger name, loop items, router branches, and error handling options. | flowId: string, stepName: string | Read |
ap_validate_flow | Validate a flow for structural issues without publishing. | flowId: string | Read |
ap_research_pieces | Research available pieces. | pieceNames?: string\[], searchQuery?: string, includeActions?: boolean, includeTriggers?: boolean | Read |
ap_search_actions | Find piece actions by describing the task in natural language (e.g. | query: string, limit?: number, pieceName?: string | Read |
ap_search_triggers | Find piece triggers (the event that starts a flow) by describing when the flow should run (e.g. | query: string, limit?: number, pieceName?: string | Read |
ap_get_piece_props | Get the detailed input property schema for a specific piece action or trigger. | pieceName: string, actionOrTriggerName: string, type: string, auth?: string, flowId?: string, input?: object | Read |
ap_resolve_property_options | Resolve dropdown options for a single piece property. | pieceName: string, actionOrTriggerName: string, type: string, propertyName: string, auth: string, input?: object, searchValue?: string | Read |
ap_resolve_property_chain | Resolve a chain of dependent dropdown properties in one call. | pieceName: string, actionOrTriggerName: string, type: string, propertyChain: array, auth: string, currentInput?: object | Read |
ap_validate_step_config | Validate a step configuration before applying it. | stepType: string, pieceName?: string, actionName?: string, triggerName?: string, input?: object, auth?: string, sourceCode?: string, packageJson?: string, loopItems?: string, settings?: object | Read |
ap_list_connections | List OAuth/app connections in the project. | pieceName?: string, displayName?: string, status?: array | Read |
ap_list_ai_models | List configured AI providers and their available models. | provider?: string | Read |
ap_list_tables | List all tables in the project with their fields (name, type, id) and row counts. | — | Read |
ap_find_records | Query records from a table with optional filtering. | tableId: string, filters?: array, limit?: number | Read |
ap_list_runs | List recent flow runs with optional filters. | flowId?: string, status?: string, limit?: number | Read |
ap_get_run | Get detailed results of a flow run including step-by-step outputs, errors, and durations. | flowRunId: string | Read |
ap_setup_guide | Get step-by-step instructions for setting up connections or AI providers. | topic: string, pieceName?: string | Read |
ap_set_project_context | Set or clear the active project context. | projectId?: string | Read |
ap_create_flow | Create a new empty flow. | flowName: string | Write |
ap_duplicate_flow | Duplicate an existing flow. | flowId: string, name?: string | Write |
ap_rename_flow | Rename an existing flow. | flowId: string, displayName: string | Write |
ap_change_flow_status | Enable or disable a flow. | flowId: string, status: string | Write |
ap_delete_flow | Permanently delete a flow and all its versions. | flowId: string | Write |
ap_lock_and_publish | Publish the current draft of a flow. | flowId: string | Write |
ap_build_flow | Create a complete flow in one call — trigger plus any number of steps. | flowName: string, trigger: object, steps: array | Write |
ap_update_trigger | Set or update the trigger for a flow. | flowId: string, pieceName: string, triggerName: string, input?: object, auth?: string, displayName?: string | Write |
ap_add_step | Add a new step to a flow. | flowId: string, parentStepName: string, stepLocationRelativeToParent: string, stepType: string, displayName: string, pieceName?: string, actionName?: string, branchIndex?: number, input?: object, auth?: string, sourceCode?: string, packageJson?: string, loopItems?: string, continueOnFailure?: boolean, retryOnFailure?: boolean | Write |
ap_update_step | Update an existing step's settings. | flowId: string, stepName: string, displayName?: string, input?: object, auth?: string, actionName?: string, loopItems?: string, sourceCode?: string, packageJson?: string, skip?: boolean, continueOnFailure?: boolean, retryOnFailure?: boolean | Write |
ap_delete_step | Delete a step from a flow. | flowId: string, stepName: string, displayName?: string | Write |
ap_add_branch | Add a conditional branch to a router step. | flowId: string, routerStepName: string, branchName: string, conditions?: array | Write |
ap_update_branch | Update the conditions and/or name of an existing router branch without affecting the steps inside it. | flowId: string, routerStepName: string, branchIndex: number, branchName?: string, conditions?: array | Write |
ap_delete_branch | Delete a branch from a router step. | flowId: string, routerStepName: string, branchIndex: number, displayName?: string | Write |
ap_manage_notes | Add, update, or delete canvas notes on a flow. | flowId: string, operation: string, noteId?: string, content?: string, color?: string, position?: object, size?: object | Write |
ap_create_table | Create a new table with an initial set of fields. | name: string, fields: array | Write |
ap_delete_table | Permanently delete a table and all its data. | tableId: string, displayName?: string | Write |
ap_manage_fields | Add, rename, or delete fields on a table. | tableId: string, operation: string, fieldId?: string, name?: string, type?: string, options?: array | Write |
ap_insert_records | Insert one or more records into a table. | tableId: string, records: array | Write |
ap_update_record | Update specific cells in a record. | tableId: string, recordId: string, fields: object | Write |
ap_delete_records | Permanently delete one or more records. | recordIds: array, displayName?: string | Write |
ap_test_flow | Test a flow end-to-end in the test environment. | flowId: string, displayName?: string, triggerTestData?: object | Exec |
ap_test_step | Test a single step within a flow. | flowId: string, stepName: string, displayName?: string, triggerTestData?: object | Exec |
ap_retry_run | Retry a failed flow run. | flowRunId: string, strategy: string | Exec |
ap_run_action | Execute a single piece action once without building or saving a flow. | pieceName: string, actionName: string, input?: object, connectionExternalId?: string | Exec |
Security & scope
- Access scope
- Network
- Sandbox
- Hosted MCP endpoint on your Activepieces instance, reached over Streamable HTTP and authenticated with OAuth — the client opens a browser on first use, so no static credential is stored. Tool categories are scoped per project, and Discovery is read-only.
- Gotchas
- Flows execute under the connections stored in Activepieces, so an agent that can publish a flow can act in every system those connections reach — Slack, databases, payment providers. Enable only the categories that project actually needs.
- Flow Building and Flow Management together let an agent publish live automation. Treat granting them as granting deploy access, not read access.
- The repository is mixed-licence: the core is MIT Expat, while packages/ee/ carries separate commercial terms. Check which applies before self-hosting a modified build.
- The Server URL is per-instance and off by default — the toggle in Settings is the enable step, and pointing a client at the wrong instance is the usual setup error.
Agent prompt pack
— copy into Claude, Cursor, or ChatGPT.Paste into Claude, Cursor, or ChatGPT. Edit the [brackets] before sending.
Recommend the best MCP servers for [task: e.g. automation & workflows work] in [client: Claude Desktop]. Constraints: - Prefer tools that are [official | open-source | read-only] — pick what matters for my use case. - Exclude MCPs that require [e.g. a paid plan, OAuth-only flows, remote-only transport]. - Return at most 3 picks, ranked. For each pick include: 1. One-sentence rationale. 2. The ready-to-paste install snippet for my client. 3. Any required secrets I need to create before installing. Cross-check the top-mcps.com listing: https://top-mcps.com/top-mcps-for-automation-workflows
Compare Activepieces against a real alternative. Swap the second MCP in [brackets] if you want a different match.
Compare Activepieces MCP vs [n8n MCP] for the following job: [describe the job, e.g. "let an agent create GitHub issues on bug triage"]. Judge them on: - Setup time and complexity (what a new user hits first). - Auth model (none / API key / OAuth 2.1) and credential risk. - Transport (stdio / Streamable HTTP / SSE) and where the server runs. - Required secrets and the blast radius if they leak. - Operational risk in an unattended agent loop. - Which one is "good enough" for a weekend prototype vs. production. End with one sentence: which should I pick for my scenario, which is: [my scenario]. References: - https://top-mcps.com/mcp/activepieces - top-mcps.com listing for n8n
Asks the agent to install and verify. Works inside Claude Code, Cursor Agent, Codex CLI.
Install the Activepieces MCP server for my [client: Claude Desktop] at the default config path for that client. Use the exact install snippet published at https://top-mcps.com/mcp/activepieces (fetch https://top-mcps.com/mcp/activepieces.json for the canonical server.json if you can read URLs). Before finishing: 1. Create the required secrets () and put them in the appropriate env block — do not hard-code them. 2. Restart or reload the client so it picks up the new server. 3. Verify the server is connected (green / running state) and at least one tool is listed. 4. If anything fails, read the client's MCP logs and report the exact error — do not silently retry. Confirm when done and list the tools the server now exposes.
Frequently asked questions
What changed
— 2 updates tracked.Refreshed install snippets and fact sheet; verified for 2026.
Initial directory listing.
More Automation & Workflows MCPs
Other tools in the same category worth evaluating.
Compared with Activepieces
Side-by-side breakdowns for the choices people most often weigh against this MCP.
Exploring Top MCPs for Automation & Workflows? See all Automation & Workflows MCPs →