Documentation Index
Fetch the complete documentation index at: https://docs.chatblocks.ai/llms.txt
Use this file to discover all available pages before exploring further.
This page is auto-generated from
apps/web/src/lib/mcp/server.ts. Do
not edit by hand — your changes will be overwritten by
pnpm --filter @chatblocks/web gen-mcp-tool-ref. CI runs
gen:check on every PR and fails if this file has drifted from the
source.39 tools
Connect with acb_live_* API key as a Bearer token. See MCP setup for client configs.
Tools marked Builder plan required are gated server-side; calling them on a Free workspace returns a JSON-RPC error with code -32603 and a PRO_REQUIRED message body.
auditLog.recent
Recent audit log. Recent audit log entries (credential decrypts, etc.) for the workspace.
Input schema
blocks.build
Compile a block’s source files via Vercel Sandbox. Compile the block’s current source files into a deployable artifact. Streams progress events (queued -> building -> succeeded/failed) and returns the final build status. On succeeded, call blocks.publish with the returned buildId to make the build live on the canvas.
Input schema
blocks.create
Builder plan required.
blocks.delete
Delete a block. Permanently delete a block and all its associated data.
Input schema
blocks.get
Get a block. Fetch a block by id, scoped to the current workspace.
Input schema
blocks.getFiles
Read block source files. Read the current source files for one block, plus latest build and published-version state. Use this before calling blocks.setFiles to see what’s there, or after blocks.build to check the build outcome.
Input schema
blocks.init
Initialize a new block. Create a new empty block with a starter file scaffold. Use this when you’re going to author the block’s source files yourself via blocks.setFiles. For one-shot AI generation, use blocks.create instead.
Input schema
blocks.list
List blocks. List blocks in the current workspace.
Input schema
blocks.publish
Publish a block build. Pin a specific successful build as the published version of a block. Pass the buildId returned by a blocks.build call that finished with status=succeeded — publishing without a build first will fail. After publish, the block’s source is served at /api/blocks/<blockId>/render and is available for canvas placement. Optional changelog is shown in version history.
Input schema
blocks.setFiles
Replace block source files. Replace the entire file set for one block. Validates the file set before writing; on validation failure returns validation.ok=false and does NOT mutate. The files array is the complete desired state (any path you omit is deleted). Pair with blocks.build to compile the new code.
Input schema
blocks.setWidgetData
Set block widget data. Hot-update the widget data for a block (bypasses the refresh pipeline).
Input schema
blocks.update
Update block metadata. Update name, description, manifest, or tags for a block. For source code changes use the editor.
Input schema
byokKeys.add
Add a BYOK AI provider key. Store a customer-supplied API key for Anthropic, OpenAI, or Google. Validated on add by pinging the provider. Encrypted at rest.
Input schema
byokKeys.delete
Delete a BYOK key. Remove a stored BYOK key. The customer’s provider account is unaffected.
Input schema
byokKeys.list
List BYOK keys. List the workspace’s stored BYOK keys (metadata only — no plaintext).
Input schema
canvases.get
Get a canvas. Fetch a canvas by id, scoped to the current workspace.
Input schema
canvases.list
List canvases. List canvases in the current workspace.
Input schema
dataSources.add
Add a data source. Connect a new data source (stripe, postgres, webhook, or mcp). Webhook type returns a one-time whsec_* secret.
Input schema
dataSources.delete
Delete a data source. Remove a data source and all associated credentials.
Input schema
dataSources.editRefreshPolicy
Edit data source refresh policy. Change the refresh cadence for a data source.
Input schema
dataSources.get
Get a data source. Fetch a data source by id, scoped to the current workspace.
Input schema
dataSources.introspectSchema
Introspect Postgres schema. List the top 50 tables in a Postgres data source, sorted by activity.
Input schema
dataSources.introspectTable
Introspect Postgres table. List columns, types, and constraints for a specific Postgres table.
Input schema
dataSources.list
List data sources. List data sources in the current workspace.
Input schema
dataSources.listMcpTools
List tools from an outbound MCP source. Connects to an mcp-typed dataSource and returns the tools it exposes. Use this before binding a block to a specific tool.
Input schema
dataSources.testConnection
Test a data source connection. Streams per-stage progress while testing a data source connection (Stripe / Postgres). Webhook and MCP types return descriptive messages.
Input schema
dataSources.triggerRefresh
Trigger a data source refresh. Schedules an immediate refresh of a data source. Returns once the refresh is queued; poll dataSources.get for lastRefreshAt.
Input schema
placements.create
Create a placement. Place a published block on a canvas at the specified coordinates.
Input schema
placements.delete
Delete a placement. Remove a block placement from its canvas.
Input schema
placements.update
Update a placement. Move or resize a placement on its canvas.
Input schema
schedule.delete
Builder plan required.
schedule.history
Get history for a scheduled run. Returns recent history rows for a scheduled run, sorted newest first.
Input schema
schedule.list
List scheduled runs. List scheduled runs in the current workspace.
Input schema
schedule.pause
Builder plan required.
schedule.preview
Builder plan required.
schedule.resume
Builder plan required.
schedule.runNow
Builder plan required.
schedule.set
Builder plan required.