Understand agent architecture: Understand how Agent Builder agents work. Learn about UI components, blueprints, state management, and how they connect to create functional agents.
Trigger blueprints via API: Enable external systems to trigger blueprints via HTTP API calls. Integrate agents with webhooks, schedulers, and third-party services programmatically.
Agent Builder Demo Application: Explore the Agent Builder demo agent that analyzes customer reviews and drafts responses. Learn how UI, blueprints, and state variables work together.
Deploy with Docker: Deploy Agent Builder agents with Docker containers. Run agents on cloud platforms, on-premises servers, or local development machines.
Using data from previous blocks: Pass data between blocks in Agent Builder blueprints. Access state variables, secrets, and results from preceding blocks with execution environment.
Create and test agents locally: Develop Agent Builder agents offline on your local machine with Writer Framework. Test agents locally before deploying to Writer Cloud.
Learn about Agent Builder: Build AI agents with a low-code visual editor. Collaborate with drag-and-drop blocks for tool calling, RAG, and custom workflows.
Python libraries installed in Agent Builder: View all Python packages available in Agent Builder including aiohttp, boto3, pandas, requests, and 50+ other libraries with version numbers.
Agent Builder Quickstart: Build your first agent in Agent Builder. Create a meeting notes summarizer with a visual interface and blueprint in this step-by-step tutorial.
Build dynamic UI components with Repeaters: Display dynamic lists of data in Agent Builder UI with Repeaters. Create repeating components for lists, dictionaries, and nested data structures.
Store secrets with Vault: Store API keys, passwords, and credentials securely with Vault in Agent Builder. Reference secrets in blueprints and Python code.
Upload, parse, and summarize PDFs: Build a PDF research paper summarizer with Agent Builder. Upload PDFs, parse content, and generate summaries for specific personas.
Use tool calling in Agent Builder: Add tool calling to Agent Builder blueprints and chatbots. Connect Knowledge Graphs and define custom functions with JSON Schema.
Understand tool calling: Learn how tool calling enables dynamic agent workflows. Understand ReAct patterns and why tool calling matters for intelligent agent orchestration.
Manage API keys: Learn how to create, manage, and delete API keys.
Application details: Retrieves detailed information for a specific no-code agent (formerly called no-code applications), including its configuration and current status.
Retrieve all jobs: Retrieve all jobs created via the async API, linked to the provided application ID (or alias).
Generate from application: Generate content from an existing no-code agent (formerly called no-code applications) with inputs.
Generate from application (async): Generate content asynchronously from an existing no-code agent (formerly called no-code applications) with inputs.
Retrieve graphs: Retrieve Knowledge Graphs associated with a no-code agent that has chat capabilities.
List applications: Retrieves a paginated list of no-code agents (formerly called no-code applications) with optional filtering and sorting capabilities.
Retry job execution: Re-triggers the async execution of a single job previously created via the Async api and terminated in error.
Associate graphs: Updates the list of Knowledge Graphs associated with a no-code chat agent.
Chat completion: Generate a chat completion based on the provided messages. The response shown below is for non-streaming. To learn about streaming responses, see the chat completion guide.
List models: Retrieve a list of available models that can be used for text generation, chat completions, and other AI tasks.
Text generation: Generate text completions using the specified model and prompt. This endpoint is useful for text generation tasks that don't require conversational context.
Delete file: Permanently delete a file from the system. This action cannot be undone.
Download file: Download the binary content of a file. The response will contain the file data in the appropriate MIME type.
Retry failed files: Retry processing of files that previously failed to process. This will re-attempt the processing of the specified files.
List files: Retrieve a paginated list of files with optional filtering by status, graph association, and file type.
Retrieve file: Retrieve detailed information about a specific file, including its metadata, status, and associated graphs.
Upload file: Upload a new file to the system. Supports various file formats including PDF, DOC, DOCX, PPT, PPTX, JPG, PNG, EML, HTML, SRT, CSV, XLS, and XLSX.
Understand rate limits: Understand API rate limits for Writer endpoints. Learn about RPM and TPM limits, best practices, and retry strategies for optimal performance.
Monitor agent performance: Monitor agent performance with session logs and observability metrics in AI Studio. Track usage, analyze conversations, and debug agent behavior.
Invoke no-code agents via the API: Turn deployed no-code agents into microservices with the Applications API. Use text generation and research agents as tools in your applications.
Run agents asynchronously: Process no-code agent requests asynchronously or in batches with the async applications API. Generate content for multiple items efficiently.
Changelog: Stay up to date with the latest changes to the Writer API and SDKs.
Generate chat completions: Create multi-turn conversations with AI using the chat completion API. Preserve conversation context and use tool calling for advanced features.
Custom connectors: Create custom connectors from OpenAPI specifications or MCP servers to extend WRITER Agent with your own APIs and tools
Choose your development path: Choose your development path for AI agents. Compare no-code builder, Agent Builder, and Writer API/SDKs for customization and technical requirements.
Encryption Key Management: Control encryption keys for data at rest with Writer-managed or customer-managed (BYOK) encryption across your organization.
Event logs: View, filter, search, and export request-level logs to debug issues and audit agent activity in AI Studio.
Add external models: Configure external AI models from providers like AWS Bedrock, Microsoft Azure, or NVIDIA NIM to use alongside Palmyra models in AI Studio
Manage files: Upload, download, list, and delete files with the File API. Use files in Knowledge Graphs and no-code agents.
Understand generative AI: Learn how generative AI works and how to leverage Writer's models, safety, and best practices for building AI-powered applications.
Configure guardrails: Configure guardrails to monitor and filter AI agent inputs and outputs for safety and compliance
Using Writer with OpenLLMetry: Monitor and debug Writer API calls with OpenLLMetry. Export traces to Jaeger, Datadog, New Relic, or Traceloop for LLM observability.
Writer AI Studio: Build, deploy, and manage AI agents with enterprise controls
Use a Knowledge Graph in a chat: Query Knowledge Graphs during chat completions with tool calling. Reference multiple graphs and configure query parameters for accurate responses.
Create and manage a Knowledge Graph: Integrate graph-based RAG into your agents with Knowledge Graph API. Achieve higher accuracy than traditional vector-based retrieval methods.
Connect your data to AI with Knowledge Graph: Understand graph-based RAG with Knowledge Graph. Learn how content ingestion, graph construction, and query processing deliver accurate AI responses.
Language support: Palmyra LLM supports 30+ languages including Arabic, French, Spanish, Hindi, and Chinese. View MMLU and BLEU benchmark scores for multilingual text generation.
Configure connectors: Configure prebuilt and custom connectors for WRITER Agent to securely access enterprise tools and your own APIs
SDKs: Install and use Writer SDKs for Python and Node.js. Get started with code examples and learn how to integrate Writer AI into your applications.
Set up BYOK encryption: Connect your own KMS from AWS, Azure, or GCP to control encryption keys for your Writer organization.
Stream responses from the API: Stream API responses in real-time with Server-Sent Events. Reduce latency and display generated content as it's created for better user experience.
Call a custom function in a chat: Extend LLM capabilities with tool calling. Execute custom functions, query databases, fetch real-time data, and integrate results into chat responses.
Web search in a chat: Search the web for real-time information during chat completions. Enable models to access current news, facts, and up-to-date data.