v0.9.7 - readme, docs
frdel committed
Nov 19, 2025 at 10:41 UTC
4c4de7ebca97eae4d14fd6bae8d8fa9995081965
2 files changed
+86
-11
README.md
+22
-8
@@ -26,14 +26,8 @@ Or see DeepWiki generated documentation:
26
27
<div align="center">
28
29
-> ### 🚨 **IMPORTANT ANNOUNCEMENT** 🚨
30
-
31
-The original GitHub and DockerHub repositories for Agent Zero have been transferred to a new namespace:
32
-
33
-- **GitHub & DockerHub:** `agent0ai/agent-zero`
34
-
35
-From now on, please use this name for both `git clone` and `docker pull` commands.
36
-
29
+> ### 🚨 **PROJECTS!** 🚨
30
+Agent Zero now supports **Projects** – isolated workspaces with their own prompts, files, memory, and secrets, so you can create dedicated setups for each use case without mixing contexts.
31
</div>
32
33
@@ -169,6 +163,26 @@ docker run -p 50001:80 agent0ai/agent-zero
163
164
## 🎯 Changelog
165
166
+### v0.9.7 - Projects
167
+[Release video](https://youtu.be/RrTDp_v9V1c)
168
+- Projects management
169
+ - Support for custom instructions
170
+ - Integration with memory, knowledge, files
171
+ - Project specific secrets
172
+- New Welcome screen/Dashboard
173
+- New Wait tool
174
+- Subordinate agent configuration override support
175
+- Support for multiple documents at once in document_query_tool
176
+- Improved context on interventions
177
+- Openrouter embedding support
178
+- Frontend components refactor and polishing
179
+- SSH metadata output fix
180
+- Support for windows powershell in local TTY utility
181
+- More efficient selective streaming for LLMs
182
+- UI output length limit improvements
183
+
184
+
185
+
186
### v0.9.6 - Memory Dashboard
187
[Release video](https://youtu.be/sizjAq2-d9s)
188
- Memory Management Dashboard
docs/extensibility.md
+64
-3
@@ -213,6 +213,11 @@ Agent Zero supports creating specialized subagents with customized behavior. The
213
- `/agents/{agent_profile}/extensions/` - for custom extensions
214
- `/agents/{agent_profile}/tools/` - for custom tools
215
- `/agents/{agent_profile}/prompts/` - for custom prompts
216
+ - `/agents/{agent_profile}/settings.json` - for agent-specific configuration overrides
217
+
218
+The `settings.json` file for an agent uses the same structure as `tmp/settings.json`, but you only need to specify the fields you want to override. Any field omitted from the agent-specific `settings.json` will continue to use the global value.
219
+
220
+This allows power users to, for example, change the AI model, context window size, or other settings for a single agent without affecting the rest of the system.
221
222
### Example Subagent Structure
223
@@ -223,15 +228,71 @@ Agent Zero supports creating specialized subagents with customized behavior. The
228
│ └── _10_example_extension.py
229
├── prompts/
230
│ └── ...
226
-└── tools/
227
- ├── example_tool.py
228
- └── response.py
231
+├── tools/
232
+│ ├── example_tool.py
233
+│ └── response.py
234
+└── settings.json
235
```
236
237
In this example:
238
- `_10_example_extension.py` is an extension that renames the agent when initialized
239
- `response.py` overrides the default response tool with custom behavior
240
- `example_tool.py` is a new tool specific to this agent
241
+- `settings.json` overrides any global settings for this specific agent (only for the fields defined in this file)
242
+
243
+## Projects
244
+
245
+Projects provide isolated workspaces for individual chats, keeping prompts, memory, knowledge, files, and secrets scoped to a specific use case.
246
+
247
+### Project Location and Structure
248
+
249
+- Projects are located under `/a0/usr/projects/`
250
+- Each project has its own subdirectory, created by users via the UI
251
+- A project can be backed up or restored by copying or downloading its entire directory
252
+
253
+Each project directory contains a hidden `.a0proj` folder with project metadata and configuration:
254
+
255
+```
256
+/a0/usr/projects/{project_name}/
257
+└── .a0proj/
258
+ ├── project.json # project metadata and settings
259
+ ├── instructions/ # additional prompt/instruction files
260
+ ├── knowledge/ # files to be imported into memory
261
+ ├── memory/ # project-specific memory storage
262
+ ├── secrets.env # sensitive variables (secrets)
263
+ └── variables.env # non-sensitive variables
264
+```
265
+
266
+### Behavior When a Project Is Active in a Chat
267
+
268
+When a project is activated for a chat:
269
+
270
+- The agent is instructed to work **inside the project directory**
271
+- Project prompts (instructions) from `.a0proj/instructions/` are **automatically injected** into the context window (all text files are imported)
272
+- Memory can be configured as **project-specific**, meaning:
273
+ - It does not mix with global memory
274
+ - The memory file is stored under `.a0proj/memory/`
275
+- Files created or modified by the agent are located within the project directory
276
+
277
+The `.a0proj/knowledge/` folder contains files that are imported into the project’s memory, enabling project-focused knowledge bases.
278
+
279
+### Secrets and Variables
280
+
281
+Each project manages its own configuration values via environment files in `.a0proj/`:
282
+
283
+- `secrets.env` – **sensitive variables**, such as API keys or passwords
284
+- `variables.env` – **non-sensitive variables**, such as configuration flags or identifiers
285
+
286
+These files allow you to keep credentials and configuration tightly scoped to a single project.
287
+
288
+### When to Use Projects
289
+
290
+Projects are the recommended way to create specialized workflows in Agent Zero when you need to:
291
+
292
+- Add specific instructions without affecting global behavior
293
+- Isolate file context, knowledge, and memory for a particular task or client
294
+- Keep passwords and other secrets scoped to a single workspace
295
+- Run multiple independent flows side by side under the same Agent Zero installation
296
297
## Best Practices
298
- Keep extensions focused on a single responsibility