Merge pull request #652 from OpenFn/adaptor-updates-mar25

Adaptor overview page updates & edits to AI Assistant links and copy

Author: taylordowns2000

adaptors/chatgpt.md

@@ -0,0 +1,43 @@
+---
+title: ChatGPT AI Adaptor
+---
+
+## About ChatGPT
+
+[ChatGPT](https://chatgpt.com/) is an AI assistant built by OpenAI. It provides
+a family of large language models (LLMs) developed for natural language
+processing tasks such as text generation, summarization, and question answering.
+
+ChatGPT models process text input to generate responses but do not have built-in memory or autonomous decision-making. They can be integrated with external tools
+or APIs to execute actions beyond text-based outputs.
+
+## Use OpenFn + LLMs for AI-Driven Workflows
+
+While LLMs (like Claude or ChatGPT) can generate insights, recommendations, or
+structured data, they often **lack direct execution capabilities**—meaning they
+need an automation system to act on their outputs.
+
+**OpenFn workflows fill this gap by orchestrating actions based on LLM-generated responses.**
+
+Remember that LLMs are imperfect. Learn more about your recommended approach to
+AI in our [Responsible AI Policy](https://www.openfn.org/ai).
+
+## Integration Options
+
+Use this adpator to integrate with the [OpenAI Platform API](https://platform.openai.com/docs/overview), to integrate ChatGPT AI models into your workflows to
+leverage their natural language processing (NLP) capabilities.
+
+Example AI-driven workflow on OpenFn: 
+![ai-driven-workflow](/img/ai-workflow-example.png)
+
+## Authentication
+
+See this adaptor's
+[Configuration docs](/adaptors/packages/chatgpt-configuration-schema) for
+technical docs on the authentication parameters required.
+
+## Helpful Links
+
+- About ChatGPT: https://chatgpt.com/
+- API docs: https://platform.openai.com/docs/overview
+- API Intro: https://www.chatbase.co/blog/chatgpt-api

adaptors/claude.md

@@ -0,0 +1,43 @@
+---
+title: Claude AI Adaptor
+---
+
+## About Claude
+
+[Claude](https://claude.ai/) is an AI assistant built by Anthropic. It provides
+a family of large language models (LLMs) developed for natural language
+processing tasks such as text generation, summarization, and question answering.
+
+Claude models process text input to generate responses but do not have built-in
+memory or autonomous decision-making. They can be integrated with external tools
+or APIs to execute actions beyond text-based outputs.
+
+## Use OpenFn + LLMs for AI-Driven Workflows
+
+While LLMs (like Claude or ChatGPT) can generate insights, recommendations, or
+structured data, they often **lack direct execution capabilities**—meaning they
+need an automation system to act on their outputs.
+
+**OpenFn workflows fill this gap by orchestrating actions based on LLM-generated responses.**
+
+Remember that LLMs are imperfect. Learn more about your recommended approach to
+AI in our [Responsible AI Policy](https://www.openfn.org/ai).
+
+## Integration Options
+
+Use this adpator to integrate with the [Claude API](https://www.anthropic.com/api), to integrate Claude AI models into your workflows to
+leverage their natural language processing (NLP) capabilities.
+
+Example AI-driven workflow on OpenFn: 
+![ai-driven-workflow](/img/ai-workflow-example.png)
+
+## Authentication
+
+See this adaptor's
+[Configuration docs](/adaptors/packages/claude-configuration-schema) for
+technical docs on the authentication parameters required.
+
+## Helpful Links
+
+- About Claude: https://claude.ai/
+- API docs: https://www.anthropic.com/api

adaptors/varo.md

@@ -0,0 +1,30 @@
+---
+title: Varo Adaptor
+---
+
+## About Varo
+
+[Varo](https://www.varo-app.org/) is a simple and free smartphone app that
+collects and forwards useful cold chain information and temperature logs to the
+email inbox of your choice.
+
+## Integration Options
+
+This is a data transformation adaptor for converting Varo data to the WHO
+Equipment Monitoring System (EMS) standard.
+
+To integrate with Varo data, you can use the relevant OpenFn adaptor (e.g.,
+Gmail) for your chosen email inbox to query the Varo emails and attachments.
+
+![example-workflow](/img/sample-cce-workflow.png)
+
+## Authentication
+
+N/A - This adaptor is for data transformation only, and not for direct
+connection to the Varo app. See note in above section.
+
+## Helpful Links
+
+- Varo website: https://www.varo-app.org/
+- WHO documentation on EMS standard:
+  https://extranet.who.int/prequal/immunization-devices/e006-temperature-monitoring-devices

docs/build/ai-assistant.md

@@ -4,23 +4,30 @@ sidebar_label: AI Assistant
 ---
 
 The OpenFn AI Assistant provides a chat interface with an AI Model to help you
-build workflows.
+build workflows. **Check out [this video](https://www.youtube.com/watch?v=3L_cGl9tWRc&ab_channel=OpenFn.org) for an introduction on how it works.**
 
 You can use it to draft, proofread and debug job code, diagnose errors, and
 understand the capabilities of the platform.
 
+:::info Build AI-Driven Workflows on OpenFn
+
+Check out OpenFn [adaptors](/adaptors) to build OpenFn workflows that orchestrate interactions with LLMs (like ChatGPT and Claude) and turn AI-powered decision-making into action and automated execution.
+
+:::
+
 ![AI Assistant](/img/ai-assistant.png)
 
-:::caution Assistant Unavailable? Can't find it? 
+:::caution Assistant Unavailable? Can't find it?
 
-For local deployments of OpenFn, the instance administrator needs to configure the AI
-Assistant before it is available. See
+For local deployments of OpenFn, the instance administrator needs to configure
+the AI Assistant before it is available. See
 [deployment docs](https://github.com/OpenFn/lightning/blob/main/DEPLOYMENT.md#ai-chat)
 here for help or contact the super user of your instance.
 
 The Assistant is available on app.openfn.org, with usage credits subject your
-project's plan. See [openfn.org/pricing](https://www.openfn.org/pricing) or contact [[email protected]](mailto:[email protected]) 
-for more information on paid plans for usage of OpenFn's cloud-hosted platform.
+project's plan. See [openfn.org/pricing](https://www.openfn.org/pricing) or
+contact [[email protected]](mailto:[email protected]) for more information on
+paid plans for usage of OpenFn's cloud-hosted platform.
 
 :::
 
@@ -36,20 +43,19 @@ versions of the assistant may be rolled out to other pages.
 All chat sessions are shared between all users of the project. You can start a
 new chat session at any time, or open an old one.
 
-Input data and run-time job logs not sent to the model. But we do include
-your step code so that the Assistant can provide a contextually relevant answer.
+Input data and run-time job logs not sent to the model. But we do include your
+step code so that the Assistant can provide a contextually relevant answer.
 
 ## A Note on Responsible AI Usage
 
 The AI assistant is built on emerging Large Language Model (terminology). Like
-other LLMS and chatbots, its capabilities are impressive, but imperfect.
+other LLMs and chatbots, its capabilities are impressive, but imperfect.
 
 Remember that ultimately, all responses are generated by an algorithm and YOU,
 the human in charge, are responsible for how its output is used. You should
 consider all responses critically and verify the output where possible.
 
-You can read more about our approach to AI in our
-[Responsible AI Policy](https://www.openfn.org/ai).
+**You can read more about our approach to AI in our [Responsible AI Policy](https://www.openfn.org/ai).**
 
 ## How to access the AI Assistant
 
@@ -68,9 +74,10 @@ You can close a chat session by clicking the `(X)` button on the top right of
 the chat interface, which will take you back to the list of sessions for this
 step.
 
-:::caution Feedback or Questions about the Assistant?  
+:::caution Feedback or Questions about the Assistant?
 
-We welcome your questions or feedback on [community.openfn.org](https://community.openfn.org/), 
-or contact [[email protected]](mailto:[email protected]) with private queries. 
+We welcome your questions or feedback on
+[community.openfn.org](https://community.openfn.org/), or contact
+[[email protected]](mailto:[email protected]) with private queries.
 
 :::

docs/build/collections.md

@@ -4,7 +4,9 @@ sidebar_label: Collections
 ---
 
 Collections provides a high-volume, high-performance storage solution built into
-OpenFn.
+OpenFn. Check out
+**[this video](https://www.youtube.com/watch?v=iXkkkzratzY&t=3s&ab_channel=OpenFn.org)**
+for an introduction.
 
 Collections is suitable for buffering, caching and aggregating data from
 Webhooks, storing large mapping files, and sharing state between workflows.
@@ -65,8 +67,8 @@ You can use the Collections API with any adaptor.
 
 :::
 
-Data is stored as key-value pairs, where the key is a unique identifier for
-some data (like a UUID, or timestamp). The value is always saved as a string
+Data is stored as key-value pairs, where the key is a unique identifier for some
+data (like a UUID, or timestamp). The value is always saved as a string
 (although you can pass JSON-compatible objects directly, which will be
 automatically serialized by the Collections API).
 
@@ -142,7 +144,9 @@ collections.each(
 );
 ```
 
-Values are uploaded to a collection through `collections.set`. All sets are "upserts" - meaning that new keys will be created for values that don't exist, and values will be updated for keys that _do_ exist. 
+Values are uploaded to a collection through `collections.set`. All sets are
+"upserts" - meaning that new keys will be created for values that don't exist,
+and values will be updated for keys that _do_ exist.
 
 The example below sets a single item:
 

docs/get-started/terminology.md

@@ -45,6 +45,13 @@ Workflows are the **"what to do"** part of automation!
 
 :::
 
+A workflow is a structured sequence of tasks, processes, or actions that are
+executed automatically based on predefined rules, triggers, and logic.
+
+When working with AI, workflows provide the structured execution needed to turn
+LLM insights into real-world actions, while AI agents enable more dynamic
+decision-making within workflows.
+
 A Workflow is a collection of a Trigger, Steps, Paths, and custom logic
 connected together to automate a specific business process or task. A Workflow
 is configured via the Canvas in the web app, or locally (via code).