Integration

Fibery integration

Fibery is a configurable work management workspace whose data lives in Databases (types) grouped in Spaces. Read the workspace schema; query Entities with Fibery's query language (fields, filters, order, limits); create, update, upsert and delete Entities and edit their collection and multi-select Fields; read and write rich text Fields and comments as Markdown or HTML; upload files from your files or a URL and get temporary links to them; list and add webhooks that report Entity changes. Each Entity tool sends one command to the commands endpoint and answers success plus result. Entities can hold personal data (users, emails). Not exposed: schema changes (Databases, Fields), views, history, the GraphQL API, deleting webhooks (use Settings, Webhooks), API token management and file downloads (they redirect; use get_file_urls). Authentication is a Fibery API token (Authorization: Token ...) plus your account name, the subdomain of your workspace URL.

Connect Fibery How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Fibery.Built from the 16 Fibery tools below, on the connection you add once.

What you can do with Fibery

Each of these is one operation on one endpoint, available both as a workflow step and as a tool an AI agent can call on its own. Nothing bundles several operations into one action, so what a run did stays readable afterwards.

Most endpoints run on the free plan. A few, mainly the ones that publish on your behalf, need a paid plan.

Entity, 5 operations

create_entityCreate entityPOSTCreates an Entity in a Database from Field values, and returns it with all its primitive, single-select and relation Fields. Collection Fields and rich text cannot be set here: use add_collection_items and update_document after.
delete_entityDelete entityPOSTDeletes one Entity of a Database, given its fibery/id.
query_entitiesQuery entitiesPOSTReads Entities of one Database with a query: which Fields to return (q/select), a filter (q/where), order (q/order-by), q/limit and q/offset. Returns a list of Entities. Results can hold personal data (users, emails).
update_entityUpdate entityPOSTUpdates primitive, single-select and relation Fields of one Entity; Fields not sent keep their values. Collection Fields use add_collection_items or set_collection_items, rich text uses update_document.
upsert_entitiesUpsert entitiesPOSTCreates a batch of Entities, unless one with the same conflict Field value exists: skip-create then skips it, update-latest updates the newest duplicate. Each Entity needs a fibery/id.

Collection item, 3 operations

add_collection_itemsAdd collection itemsPOSTAdds existing Entities (or multi-select options, or uploaded Files) to a collection Field of an Entity; items already in the collection are ignored.
remove_collection_itemsRemove collection itemsPOSTRemoves Entities (or multi-select options) from a collection Field of an Entity; items not in the collection are ignored.
set_collection_itemsSet collection itemsPOSTReplaces the whole content of a collection Field of an Entity: every current item is removed and the given Entities (or multi-select options) are added.

Document, 2 operations

get_documentGet documentGETReturns the content of a rich text Field or comment (a collaborative document) given its secret, as Markdown (default), HTML, JSON or plain text.
update_documentUpdate documentPUTSets the whole content of a rich text Field or comment (a collaborative document), given its secret, from Markdown (default), HTML or JSON. A 200 answer means the update succeeded.

File, 2 operations

upload_fileUpload filePOSTUploads a file into Fibery storage and creates its File Entity; returns fibery/id (attach it with add_collection_items on a Files Field) and fibery/secret (for get_file_urls).
upload_file_from_urlUpload file from urlPOSTHas Fibery download a file from a URL into its storage and create its File Entity; returns fibery/id (to attach the file) and fibery/secret. Optional method and headers reach the source server.

Webhook, 2 operations

create_webhookCreate webhookPOSTAdds a webhook: on Entity changes in one Database, Fibery POSTs the new and previous values to your URL. Rich text changes are not sent. Needs Architect access to the Space or Database.
list_webhooksList webhooksGETLists the workspace webhooks with the URL each one calls, the Database it watches, its state and its latest 50 runs (HTTP status, time taken, request time).

File url, 1 operation

get_file_urlsGet file urlsPOSTReturns temporary public URLs, valid for 60 minutes, for stored files given their fibery/secret values (from upload_file or a query on a Files Field).

Schema, 1 operation

get_schemaGet schemaPOSTReturns the workspace schema: every Database (fibery/types) with its Fields, their types and relation or collection flags. Use it to learn the Space/Name of Databases and Fields before a query. Can be a few hundred KB.

Use Fibery from Claude Code, Cursor or Codex

Connect Fibery in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Fibery operations become tools the client can call, next to every other integration you connected.

Claude Code

claude mcp add --transport http livecontext https://livecontext.ai/mcp --header "X-API-Key: YOUR_API_KEY"

Codex (~/.codex/config.toml)

[mcp_servers.livecontext]
url = "https://livecontext.ai/mcp"

[mcp_servers.livecontext.http_headers]
"X-API-Key" = "YOUR_API_KEY"

Cursor, Claude Desktop and other JSON configs

{
  "mcpServers": {
    "livecontext": {
      "type": "http",
      "url": "https://livecontext.ai/mcp",
      "headers": {
        "X-API-Key": "YOUR_API_KEY"
      }
    }
  }
}

Set up the MCP server step by step

How to connect Fibery

Create your workspace
Sign up free. Nothing to install.
Connect Fibery once
Add your Fibery connection in Settings. Every tool from this integration reuses it, and secrets are encrypted at rest and visible only to you.
Describe the job in chat
The workflow builds itself in front of you and picks the endpoints it needs. Drag the steps yourself on the canvas whenever you would rather be explicit.