img fix + knowledge_tool
- A knowledge_tool in-depth explanation was missing from docs/architecture.md#tools - Other fixes
Alessandro committed
Oct 15, 2024 at 16:13 UTC
24542faf676403c3a376846d0d678ccfa6652eca
5 files changed
+28
-12
README.md
+15
-11
@@ -55,7 +55,7 @@
55
- No coding is required, only prompting and communication skills.
56
- With a solid system prompt, the framework is reliable even with small models, including precise tool usage.
57
58
-
58
+
59
60
## Keep in mind
61
1. **Agent Zero can be dangerous!**
@@ -85,17 +85,26 @@ If you cannot provide all the necessary conditions or API keys, just change the
85

86
87
## Setup
88
-A detailed setup guide with a video can be found in /docs/installation.md. Open the [Documentation](docs/installation.md#in-depth-guide-for-windows-and-macos).
88
+A detailed setup guide for Windows, macOS and Linux with a video can be found in the new Agent Zero Documentation at [this page](docs/installation.md#in-depth-guide-for-windows-and-macos).
89
90
-## Changelog [since version 0.7]
90
+## Consult the Documentation
91
+The documentation dives deep into the framework and its features. It is a good place to start if you are new to Agent Zero. Click [here](docs/README.md) to see the Documentation.
92
92
-### v0.7.1
93
+## Coming up
94
+- **Preinstalled bundles**
95
+- **Knowledge tool open-sourcing and web scraping tool**
96
+- **User interaction refinements**
97
+- **In-context switchable LLMs**
98
+
99
+### Changelog [since version 0.7]
100
+
101
+#### v0.7.1
102
- **Bug Fixes**
103
- **Persistent Chats** - Serialized to /tmp/chats and automatically loaded in run_ui.py on startup
104
- **Preinstalled binaries and bundler scripts**
105
- **Documentation stack merged into the repository**
106
98
-### New in version 0.7
107
+#### New in version 0.7
108
> - UI Revamp
109
> - Auto Memory
110
> - Instruments
@@ -105,9 +114,4 @@ A detailed setup guide with a video can be found in /docs/installation.md. Open
114
> [!NOTE]
115
> **Changes to launch files since v0.6:**
116
> - main.py file has been replaced with run_ui.py (webui) and run_cli.py (terminal) launch files.
108
-> - configuration has been moved to initialize.py for both webui and terminal launch files.
109
-
110
-## Coming up
111
-- **Preinstalled bundles**
112
-- **Knowledge tool improvements and web scraping tool**
113
-- **User interaction refinements**
\ No newline at end of file
117
+> - configuration has been moved to initialize.py for both webui and terminal launch files.
\ No newline at end of file
docs/architecture.md
+12
@@ -69,6 +69,18 @@ The memory system is a critical component of Agent Zero, enabling the agent to l
69
| memory_tool | Enables agents to save, load, delete and forget information from memory. |
70
| webpage_content_tool | Enables agents to fetch and analyze the text content of webpages. |
71
72
+- **Knowledge Tool:** The `knowledge_tool` uses DuckDuckGo and Perplexity APIs to search the web and retrieve information. It can also search the local knowledge base and memory for relevant information. The tool returns a summary of the information, which can be used by the agent to make decisions or answer questions.
73
+
74
+> [!TIP]
75
+> The Perplexity API key is the only service recommended for a full experience of
76
+> Agent Zero, but you can use DuckDuckGo by leaving the Perplexity API key value
77
+> empty. DuckDuckGo usage in Agent Zero is free of charge but has its rate limits.
78
+
79
+> [!NOTE]
80
+> It's important to note that the DuckDuckGo API is not as powerful as Perplexity
81
+> and may not return accurate or relevant results for some queries. The Perplexity
82
+> API is yet to be replaced with an open-source alternative.
83
+
84
- **Custom Tools:** Users can create custom tools to extend Agent Zero's capabilities. Custom tools can be integrated into the framework by defining a tool specification, which includes the tool's prompt to be placed in `/prompts/$FOLDERNAME/agent.system.tool.$TOOLNAME.md`, as further detailed [here](#adding-tools).
85
86
- **Tools vs. Instruments:** Tools are always present in system prompt, so you should keep them to minimum. To save yourself some tokens, use the [Instruments module](#adding-instruments) to call custom scripts or functions.
docs/res/ui_screen.png
Binary files /dev/null and b/docs/res/ui_screen.png differ
docs/res/ui_screen1.png
Binary files a/docs/res/ui_screen1.png and /dev/null differ
docs/usage.md
+1
-1
@@ -74,7 +74,7 @@ app.run(request_handler=NoRequestLoggingWSGIRequestHandler, host="0.0.0.0", port
74
75
## Using code_execution_tool outside of the Docker Container
76
> [!CAUTION]
77
-> VERY IMPORTANT NOTICE: We believe in safe and ethical AI development, and highly
77
+> We believe in safe and ethical AI development, and highly
78
> suggest you to use the Docker container designed for running Agent Zero.
79
> This is a dangerous and untested feature, and we are not responsible for any damage
80
> or illegal activities or legal liabilities caused by the use of this feature.