Integration

Writer integration

Writer is an enterprise generative AI platform built on its Palmyra models. Generate chat completions and text with Palmyra models (palmyra-x6 is current; palmyra-x5 and palmyra-x4 are deprecated by Writer on 2026-12-14), optionally grounded with a built-in Knowledge Graph, web search, translation or another model as a tool, run the no-code agents deployed in AI Studio synchronously or as async jobs, build Knowledge Graphs from uploaded files or web URLs and ask them questions, upload, download and manage files, and parse PDFs to text or Markdown. Calls are billed to your Writer account: generation per token at the model's rate, PDF parsing and Knowledge Graph file parsing per page, Knowledge Graph storage per day. Left out: streaming responses, the deprecated web search endpoint and the deprecated Palmyra Vision endpoint (send images in a chat message to palmyra-x5 instead). Authentication is an API key from AI Studio (Admin settings, API Keys), sent as a Bearer token; the key's API agent decides which capabilities it can call.

Connect Writer How integrations work

Ask it in chat

Say what should happen.
The workflow builds itself.

When this happens, do that in Writer.Built from the 27 Writer tools below, on the connection you add once.

What you can do with Writer

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.

Graph, 7 operations

create_graphCreate graphPOSTCreates a Knowledge Graph and returns its id. It is org-wide unless team_ids is given (with an org-scoped key); a team-scoped API key always assigns it to its own team. Add content with add_file_to_graph or update_graph urls.
delete_graphDelete graphDELETEDeletes a Knowledge Graph and returns its id with deleted true.
get_graphGet graphGETReturns one Knowledge Graph: name, description, type, web connector URLs, team ids and how many of its files are in progress, completed or failed.
list_application_graphsList application graphsGETReturns the ids of the Knowledge Graphs connected to a no-code chat agent.
list_graphsList graphsGETLists Knowledge Graphs with their file processing counts. By default only org-wide graphs are returned; pass team_ids to include graphs deployed to those teams. A team-scoped API key always sees only its team's graphs.
set_application_graphsSet application graphsPUTReplaces the list of Knowledge Graphs connected to a no-code chat agent (only chat agents support Knowledge Graphs) and returns the new list.
update_graphUpdate graphPUTUpdate the name, description, web connector URLs, or team assignment of a Knowledge Graph.

File, 6 operations

add_file_to_graphAdd file to graphPOSTAdds an uploaded file to a Knowledge Graph and returns the file with its graph_ids and processing status. The file inherits the graph's team access.
delete_fileDelete fileDELETEPermanently delete a file from the system. This action cannot be undone.
get_fileGet fileGETRetrieve detailed information about a specific file, including its metadata, status, and associated graphs.
list_filesList filesGETRetrieve a paginated list of files with optional filtering by status, graph association, and file type.
remove_file_from_graphRemove file from graphDELETERemoves a file from a Knowledge Graph and returns the file id with deleted true.
upload_fileUpload filePOSTUploads a file (PDF, DOC, DOCX, PPT, PPTX, JPG, PNG, EML, HTML, SRT, CSV, XLS, XLSX, TXT and more) to Writer and returns its id and processing status. Optionally adds it to a Knowledge Graph with graphId.

Application, 3 operations

generate_from_applicationGenerate from applicationPOSTRuns a deployed no-code agent with the given inputs and waits for its output: title (the output field name) and suggestion (the generated content).
get_applicationGet applicationGETReturns one deployed no-code agent with its type, status and the input fields it expects (name, input_type, required, options); generate_from_application takes those inputs.
list_applicationsList applicationsGETRetrieves a paginated list of no-code agents (formerly called no-code applications) with optional filtering and sorting capabilities.

Job, 3 operations

create_application_jobCreate application jobPOSTStarts an async job that runs a no-code agent with the given inputs and returns the job id with status in_progress at once. Read the result later with get_application_job (data.suggestion once completed).
get_application_jobGet application jobGETReturns one async no-code agent job: its status (in_progress, completed or failed), and the generated output in data.suggestion once completed, or error when it failed.
list_application_jobsList application jobsGETLists the async jobs created for a no-code agent, with each job's status (in_progress, completed or failed), timestamps and, when completed, its result.

Retry, 2 operations

retry_application_jobRetry application jobPOSTRuns a failed async job again and returns the job id with its new status.
retry_failed_filesRetry failed filesPOSTRe-runs processing for files whose status is failed (see list_files with status=failed).

Chat, 1 operation

create_chat_completionCreate chat completionPOSTGenerates the next assistant message for a conversation with a Palmyra model (or an external model configured in AI Studio). The reply is in choices[0].message.content, with token usage.

Completion, 1 operation

generate_textGenerate textPOSTGenerates text from a single prompt with a Palmyra model, without conversation history. The generated text is in choices[0].text.

Download, 1 operation

download_fileDownload fileGETDownloads the original content of an uploaded file and returns it as a stored file reference.

Model, 1 operation

list_modelsList modelsGETLists the models available to your organization's API key, with each model's id (the value to pass as model to create_chat_completion and generate_text) and name.

Pdf parser, 1 operation

parse_pdfParse pdfPOSTExtracts the content of an uploaded PDF file as plain text or Markdown and returns it in content. Writer bills it per page.

Question, 1 operation

ask_graph_questionAsk graph questionPOSTAsks a question to one or more Knowledge Graphs and returns the answer with the source snippets (file id and text) and, when requested, the generated subqueries and their answers.

Use Writer from Claude Code, Cursor or Codex

Connect Writer in LiveContext, create an API key in Settings, then add the LiveContext MCP server to your client. Its Writer 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 Writer

Create your workspace
Sign up free. Nothing to install.
Connect Writer once
Add your Writer 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.