From cd5911c53ba8694b55cdafa323e111cbb22b2534 Mon Sep 17 00:00:00 2001 From: Braelyn Boynton Date: Mon, 27 Jan 2025 13:35:25 -0800 Subject: [PATCH] readme and docs for 0.3 --- README.md | 35 ++++++++++++++++++++------- docs/essentials/agentops.mdx | 4 +-- docs/essentials/generating-agents.mdx | 10 +++++--- docs/essentials/generating-tasks.mdx | 9 ++++--- docs/quickstart.mdx | 4 +-- 5 files changed, 40 insertions(+), 22 deletions(-) diff --git a/README.md b/README.md index af37b926..91debd78 100644 --- a/README.md +++ b/README.md @@ -30,6 +30,13 @@ AgentStack scaffolds your _agent stack_ - the tech stack that collectively is yo agentstack init

+### Currently Supported Providers +- **LLMs**: Most all notable LLMs and providers are supported via LiteLLM or LangChain +- **Framework**: Currently supported frameworks include CrewAI and LangGraph + - Roadmap: Pydantic AI, Eliza, AG2 and Autogen +- **Tools**: Maintaining the largest repository of framework-agnostic tools! All tools listed [here](https://docs.agentstack.sh/tools/community) +- **Observability**: AgentOps baked in by default with first-tier support + ### Get Started Immediately You **don't** need to install or configure tools like LangChain or LlamaIndex.
@@ -44,7 +51,7 @@ Create a project, and you're good to go. To create a new agent project, run: ```sh -pip install agentstack +uv pip install agentstack agentstack init ``` @@ -56,10 +63,17 @@ Once the initialization is done, you can open your project folder: ```sh cd +uv pip install +``` + +then run the default agent! + +```sh +agentstack run ``` ### Templates -Additionally, pre-built templates are available as a shortcut to building your project. [View the available templates](). +Additionally, pre-built templates are available as a shortcut to building your project. [View the available templates](https://docs.agentstack.sh/templates). ## Building Agent Functionality @@ -101,26 +115,29 @@ Runs the agent project in development mode.
- **No Lock-In:** You can customize your setup at any time. AgentStack is designed to make it easy to get the components you need running right off the bat; it's up to you what to do next. +- **Provider Agnosticism:** AgentStack is designed to be provider agnostic in all ways. This means especially that the framework layer of your agent stack should be interoperable with every other layer. Frameworks must still be supported by AgentStack. See supported frameworks above. + AgentStack is not designed to be a low-code solution to building agents. Instead it is a great head-start for starting an agent project from scratch. ## Roadmap -### Framework Agnosticism - -Development of AgentStack is being done primarily on [CrewAI](https://crewai.com). We will soon be working to make AgentStack fully framework-agnostic, meaning that any supported multi-agent framework can be used for your project. - ### Tools -- Core Tools built by AgentStack +- More [core tools](https://docs.agentstack.sh/tools/core) built by AgentStack - Preferred partners in the package directly - Community partner tools added through external repositories +### New Stack Layers +- Prompting layer (DSPy) +- Eval Layer +- UI layer + ### Other Features - Generated testing - Integrated benchmarking - Easy integration of tools for browsing, RAG, and more. - A fast interactive test runner with built-in support for coverage reporting. -- A live development server that warns about common mistakes. -- A build script to bundle your project for production. +- A live development server that warns about common mistakes (in progress). +- A build script to bundle your project for production (in progress). ## License diff --git a/docs/essentials/agentops.mdx b/docs/essentials/agentops.mdx index 2ba5e1e7..1a1beca6 100644 --- a/docs/essentials/agentops.mdx +++ b/docs/essentials/agentops.mdx @@ -8,7 +8,7 @@ description: 'Using AgentOps makes the agent development process manageable' AgentOps provides monitoring for agents in development and production. It provides a dashboard for tracking agent performance, session replays, and custom reporting. -Additionally, AgentOps provides session drilldowns for viewing Crew agent interactions, LLM calls, and tool usage in real-time. +Additionally, AgentOps provides session drilldowns for viewing agent interactions, LLM calls, and tool usage in real-time. This feature is useful for debugging and understanding how agents interact with users as well as other agents. ![Overview](https://github.com/AgentOps-AI/agentops/raw/main/docs/images/external/app_screenshots/session-replay.png) @@ -44,6 +44,6 @@ AgentOps is baked into AgentStack projects by default! You only need to add an A ## Further Information -To get started, create an [AgentOps account](https://agentops.ai/?=crew). +To get started, create an [AgentOps account](https://agentops.ai/?=agentstack). For feature requests or bug reports, please reach out to the AgentOps team on the [AgentOps Repo](https://github.com/AgentOps-AI/agentops). \ No newline at end of file diff --git a/docs/essentials/generating-agents.mdx b/docs/essentials/generating-agents.mdx index dbdc0ec4..7c685eb3 100644 --- a/docs/essentials/generating-agents.mdx +++ b/docs/essentials/generating-agents.mdx @@ -9,11 +9,13 @@ To generate a new agent for your project, run: agentstack generate agent ``` -This command will modify two files, `crew.py` and `agents.yaml`. +This command will modify two files, your agent file (`crew.py` / `graph.py`) and `agents.yaml`. -## crew.py +## your agent file -This is the file that declares each of your agents and tasks. It's the core of your AgentStack project with Crew. +This is the file that declares each of your agents and tasks. It's the core of your AgentStack project and how AgentStack configures your framework. +- Crew projects have `crew.py` +- LangGraph projects have `graph.py` ## agents.yaml @@ -26,6 +28,6 @@ Each agent has three prompt params: - Backstory And one configuration param: -- LLM - This value tells crew which model to use for this specific agent. It must be in `/` format +- LLM - This value tells AgentStack which model to use for this specific agent. It must be in `/` format Ex: `openai/gpt-4o` \ No newline at end of file diff --git a/docs/essentials/generating-tasks.mdx b/docs/essentials/generating-tasks.mdx index ef548dcc..fbb23f67 100644 --- a/docs/essentials/generating-tasks.mdx +++ b/docs/essentials/generating-tasks.mdx @@ -9,12 +9,13 @@ To generate a new task for your project, run: agentstack generate task ``` -This command will modify two files, `crew.py` and `agents.yaml`. +This command will modify two files, your agent file (`crew.py`/`graph.py`) and `agents.yaml`. -## crew.py - -This is the file that declares each of your agents and tasks. It's the core of your AgentStack project with Crew. +## your agent file +This is the file that declares each of your agents and tasks. It's the core of your AgentStack project and how AgentStack configures your framework. +- Crew projects have `crew.py` +- LangGraph projects have `graph.py` ## agents.yaml diff --git a/docs/quickstart.mdx b/docs/quickstart.mdx index 85dc3403..7878e51c 100644 --- a/docs/quickstart.mdx +++ b/docs/quickstart.mdx @@ -33,9 +33,7 @@ agentstack init --template= ## Building your project -AgentStack 0.2 is built on top of [CrewAI](https://docs.crewai.com/introduction). The majority of your development experience will be with Crew. - -When working with your Crew project, AgentStack makes generating new agents and tasks easy. +AgentStack 0.3 and beyond is framework-agnostic! Choose any supported framework and start building. We help you scaffold your project quickly, but building complex routing and workflows is still done at the framework level. Be sure to understand how to build with your framework and turn to their documentation for guidance. ### Agents