main
md 435 lines 14.4 KB
Rendered Raw
1 # Usage Guide
2
3 This guide is the practical tour of Agent Zero after installation. It explains
4 what you can do in the Web UI, what to try first, and where to go when you want
5 the deeper source-linked explanation.
6
7 For architecture, backend flow, Web UI internals, plugin lifecycle, and API
8 details, use [DeepWiki for Agent Zero](https://deepwiki.com/agent0ai/agent-zero).
9
10 ![Agent Zero first task](../res/usage/first-task.png)
11
12 ## Basic Operations
13
14 Agent Zero is built around a chat, a working Linux environment, and a Web UI that
15 lets you watch and steer the work.
16
17 Common places to start:
18
19 - **New Chat:** start a clean conversation.
20 - **Projects:** give a chat its own workspace, files, memory, secrets, and instructions.
21 - **Memory:** review what Agent Zero has learned or imported.
22 - **Tasks:** create scheduled, planned, or manual automations.
23 - **Files:** open the Agent Zero file browser.
24 - **Settings:** configure models, credentials, preferences, plugins, and backup.
25 - **Browser:** open the live Browser surface when you want to watch browsing or annotate a page.
26 - **Desktop:** open the live Linux desktop when you want GUI apps, a terminal window, or LibreOffice Cowork.
27
28 ![Dashboard actions](../res/usage/webui/dashboard.png)
29
30 The chat input also has action buttons for attachments, pausing, nudging, compacting,
31 and opening helpful views such as context or history.
32
33 ![Action buttons](../res/usage/action-btns.png)
34
35 Use **Restart** from the sidebar when you need the framework to reload after
36 settings or code changes.
37
38 ## Plugins And Plugin Hub
39
40 Plugins add integrations, tools, panels, and automation helpers.
41
42 Open **Plugins** from the dashboard or sidebar to see what is installed.
43
44 ![Plugins](../res/usage/plugins/plugins-list-01.png)
45
46 Use the **Browse** tab or **Install** button to open the Plugin Hub.
47
48 ![Plugin Hub](../res/usage/plugins/plugin-hub-main-view.png)
49
50 Before installing a plugin, read its description, README, permissions, and source
51 link. Treat plugins like any other code you run in your workspace: install the
52 ones you trust and remove what you do not use.
53
54 When you want to make your own first plugin, start with something small and
55 visible. The [Create a Small Plugin](create-plugin.md) guide walks through a
56 local Web UI plugin that adds an unread dot to the chat list and then reviews it
57 with `a0-review-plugin`.
58
59 ## Skills, Agent Profiles, And Model Presets
60
61 The small controls around the chat input let you shape the current conversation
62 without opening the full Settings screen.
63
64 ### Skills
65
66 Skills are focused instructions Agent Zero can load when it needs them. You can
67 also pin a skill manually for the current chat.
68
69 Click the **+** button in the chat input, then click **Skills**.
70
71 ![Open Skills from chat input](../res/usage/webui/chat-more-actions-skills.png)
72
73 Use the selector to add or remove active skills.
74
75 ![Skills selector](../res/usage/webui/skills-selector-checked.png)
76
77 Active skills are added to the **Protocol** part of the prompt, so keep the
78 list short and intentional. See the [Skills guide](skills.md).
79
80 ### Agent Profiles
81
82 Agent Profiles change the role, tone, and prompt instructions for the selected
83 chat.
84
85 ![Agent Profile selector](../res/usage/webui/agent-profile-selector.png)
86
87 Use the profile menu near the chat input to switch the current chat or edit a
88 profile. Select **Manage agents** to create, duplicate, reset, delete, or change
89 profile availability for Global or one project.
90
91 ![Manage agents with a project-specific availability choice](../res/usage/webui/agent-profile-manager.png)
92
93 Use **Settings -> Agent Config** when you want to change the default profile for
94 new Global chats.
95
96 See the [Agent Profiles guide](agent-profiles.md).
97
98 ### Model Presets
99
100 Model Presets are named shortcuts for model choices. Use them for setups like
101 `Best`, `Balanced`, `Fast Cheap`, or a model name you can spot quickly.
102
103 ![Model preset selector](../res/usage/webui/model-preset-selector.png)
104
105 Click **Edit presets** when you want to add or rename presets.
106
107 ![Model presets editor](../res/usage/webui/model-presets-editor.png)
108
109 See the [Model Presets guide](model-presets.md).
110
111 ## File Attachments
112
113 Attach files when the agent should read, summarize, transform, or organize them.
114
115 ![File attachments](../res/usage/attachments-1.png)
116
117 You can attach one file or several files, then describe what should happen:
118
119 ```text
120 Read these PDFs and create a short comparison table.
121 ```
122
123 ```text
124 Move these files into a clean folder structure and explain what changed.
125 ```
126
127 Attached files are visible in the chat input before you send the message, so you
128 can remove mistakes before Agent Zero starts working.
129
130 ## Tool Usage
131
132 You usually do not need to name tools. Say what you want done and Agent Zero will
133 choose whether it needs the browser, code execution, files, knowledge, plugins,
134 or another available capability.
135
136 Good prompts are specific about the desired result:
137
138 ```text
139 Research three deployment options for this app. Cite sources and finish with a recommendation.
140 ```
141
142 ```text
143 Open the attached CSV, find the main trend, and create a chart I can edit later.
144 ```
145
146 ```text
147 Inspect this repository and propose the safest first improvement before changing files.
148 ```
149
150 When you do want internals, use
151 [DeepWiki for Agent Zero](https://deepwiki.com/agent0ai/agent-zero).
152
153 ### Browser Tool And Surface
154
155 The Browser has two parts:
156
157 - the `browser` tool, which the agent can call directly;
158 - the visible Browser surface in the Canvas, where you can watch and annotate pages.
159
160 ![Browser Canvas and tool history](../res/usage/browser/browser-canvas-wide.png)
161
162 Ask naturally:
163
164 ```text
165 Use the Browser tool to compare these pages and take screenshots of the important parts.
166 ```
167
168 ```text
169 Open my local app in the Browser. I will annotate the page, then you can fix the issues.
170 ```
171
172 For screenshots, history, annotations, Docker browser mode, host-browser mode
173 through A0 CLI, privacy controls, and Chrome extensions, see the
174 [Browser Guide](browser.md).
175
176 External browser MCP tools are still useful for specialized setups. See
177 [MCP Setup](mcp-setup.md).
178
179 ### Desktop Surface
180
181 The Desktop surface opens Agent Zero's own Linux desktop in the Canvas.
182 Use it when you want the agent to work visually with GUI apps, open a terminal,
183 or cowork with you in LibreOffice.
184
185 ![Desktop Canvas](../res/usage/webui/desktop-canvas.png)
186
187 The **New** menu can create Markdown, Writer, Spreadsheet, and Presentation
188 files. Writer, Calc, and Impress run inside the Desktop, so you can edit by hand
189 while Agent Zero creates, updates, saves, and verifies the same files.
190
191 For the screenshot walkthrough and prompt examples, see the
192 [Desktop Guide](desktop.md).
193
194 ### Agent-To-Agent Communication
195
196 Agent Zero instances can communicate through A2A when you want multiple
197 instances to collaborate.
198
199 Use A2A when you have a clear reason to split work across Agent Zero instances,
200 such as a specialist server, a remote machine, or a project-specific agent. See
201 [A2A Setup](a2a-setup.md).
202
203 ### Multi-Agent Cooperation
204
205 Inside a single Agent Zero instance, the main agent can create subordinate agents
206 to investigate focused parts of a larger job.
207
208 ![Multi-agent cooperation](../res/usage/multi-agent.png)
209
210 This is useful for research, code review, comparison work, and tasks where one
211 agent should gather information while another keeps the main plan moving.
212
213 ## Projects
214
215 Projects tell Agent Zero what world it is working in. Use one when a chat needs
216 its own files, instructions, memory, secrets, or model settings.
217
218 ![Project active in chat](../res/usage/webui/project-active-chat.png)
219
220 The simple flow:
221
222 1. Open **Projects** from the dashboard.
223 2. Click **Create project**.
224 3. Give it a clear title.
225 4. Add a short description.
226 5. Write practical instructions.
227 6. Save it.
228 7. Open a chat and choose the project from the top-right project picker.
229
230 ![Project instructions](../res/usage/webui/project-instructions-filled.png)
231
232 Good project instructions tell Agent Zero what should be different in that
233 workspace:
234
235 ```markdown
236 When this project is active:
237 - Explain steps in plain language before technical detail.
238 - Prefer screenshots, checklists, and concrete examples.
239 - Keep generated files inside this project unless I ask otherwise.
240 - Ask before using credentials, private data, or external accounts.
241 ```
242
243 Use projects for client work, code repositories, research topics, recurring
244 reports, and any workflow where context matters.
245
246 See the [Projects guide](projects.md) for the full screenshot walkthrough.
247
248 ## Tasks And Scheduling
249
250 Tasks let Agent Zero run work later, repeatedly, or on demand.
251
252 Use tasks for:
253
254 - morning or weekly reports;
255 - monitoring a source and summarizing changes;
256 - recurring cleanup or export jobs;
257 - project-specific checks;
258 - manual batch jobs you want to run again.
259
260 Open **Tasks** from the dashboard or sidebar.
261
262 ![Task scheduler](../res/usage/tasks/scheduler-1.png)
263
264 When creating a task, focus on four things:
265
266 - **Name:** what you will recognize later.
267 - **Type:** scheduled, planned, or ad-hoc.
268 - **Project:** optional, but recommended when the task needs specific context or secrets.
269 - **Prompt:** the actual work Agent Zero should perform.
270
271 ![Edit task](../res/usage/tasks/edit-task.png)
272
273 Example:
274
275 ```text
276 Name: Weekly docs review
277 Type: Scheduled
278 Project: Documentation
279 Prompt: Check the docs project for stale screenshots, broken links, and confusing sections. Summarize what needs attention.
280 ```
281
282 Project-scoped tasks inherit project instructions, variables, secrets, files, and
283 memory. That means you can improve task behavior later by improving the project
284 instead of repeating every rule in every task.
285
286 ## Secrets And Variables
287
288 Use **Secrets** for sensitive values such as API keys, tokens, passwords, and
289 credentials.
290
291 Use **Variables** for non-sensitive settings such as regions, URLs, usernames,
292 formats, or feature flags.
293
294 Refer to them by name in chat:
295
296 ```text
297 Use the project GITHUB_TOKEN to check repository status.
298 ```
299
300 Do not paste credentials into chat messages or public files. Keep your own copy
301 of important secrets because backups may not include every secret.
302
303 ## Remote Access Via Tunneling
304
305 Tunnels let you reach your local Agent Zero instance from another device or
306 share it temporarily.
307
308 Before creating a tunnel:
309
310 - set UI authentication;
311 - understand that anyone with the tunnel URL can try to open your instance;
312 - stop the tunnel when you no longer need it.
313
314 Open **Settings -> External Services -> Flare Tunnel** to create or stop a tunnel.
315
316 ## Voice Interface
317
318 Agent Zero supports text-to-speech and speech-to-text through built-in voice plugins:
319
320 - `_kokoro_tts` provides container-side Kokoro speech synthesis when enabled.
321 - `_whisper_stt` provides local Whisper transcription and adds the microphone control when enabled.
322 - Browser-native `speechSynthesis` remains the fallback output path when `_kokoro_tts` is disabled.
323
324 Use the **Voice** section in Agent settings or the plugin settings in **Agent Plugins** to configure providers. Use the sidebar **Speech** preference when you want Agent Zero to read responses automatically.
325
326 Use speech when you want to listen while doing something else, dictate a prompt,
327 or make the interface more accessible.
328
329 ![Text to speech controls](../res/usage/ui-tts-stop-speech1.png)
330
331 Speech-to-text settings live in the Whisper STT plugin card and include model size, language code, voice message handling, silence threshold, and recording behavior. The microphone button appears in the chat input when `_whisper_stt` is enabled.
332
333 ![Speech to text settings](../res/usage/ui-settings-5-speech-to-text.png)
334
335 > [!IMPORTANT]
336 > Whisper STT and Kokoro TTS operate locally within the Docker/container runtime when their plugins are enabled.
337 > Browser fallback TTS runs locally in the browser. No voice path requires OpenAI APIs.
338
339 ## Mathematical Expressions
340
341 Agent Zero can render mathematical notation with KaTeX.
342
343 ![KaTeX display](../res/usage/ui-katex-2.png)
344
345 Ask for the format you want:
346
347 ```text
348 Solve this step by step and show the final equations in KaTeX.
349 ```
350
351 ## File Browser
352
353 The File Browser lets you inspect and manage files inside the Agent Zero
354 environment.
355
356 ![File Browser](../res/usage/file-browser.png)
357
358 Use it to:
359
360 - upload files;
361 - download generated work;
362 - create folders;
363 - rename or delete files;
364 - open editable text files;
365 - inspect the project or `/a0/usr` workspace.
366
367 For file-based work, prefer `/a0/usr` or a project workspace. Avoid storing
368 important work only in temporary directories.
369
370 ## Memory Management
371
372 Memory is where Agent Zero keeps useful remembered information from conversations
373 and imported knowledge. It is powerful, but it is not magic. Long-term AI memory
374 still needs curation; this is not fully solved even by the largest AI labs and
375 companies.
376
377 Open **Memory** when you want to search, review, edit, copy, or remove stored
378 entries.
379
380 ![Memory dashboard](../res/usage/memory-dashboard.png)
381
382 The controls let you choose a memory directory, filter by area, set a result
383 limit, search, adjust match threshold, and clear filtered results.
384
385 ![Memory dashboard controls](../res/usage/webui/memory-dashboard-controls.png)
386
387 Use memory deliberately:
388
389 - keep durable facts and useful patterns;
390 - remove old test data;
391 - edit memories that became inaccurate;
392 - use project memory for project-specific context;
393 - create a backup before large cleanup.
394
395 If Agent Zero does something unexpected, repeats a wrong behavior, or seems to
396 remember the wrong thing, Memory is one of the first places to look. A stale or
397 incorrect memory can poison the processing instead of helping it.
398
399 Click a memory row to inspect its full content and metadata. From the detail
400 view you can copy, edit, or delete the entry.
401
402 ![Memory editing](../res/usage/memory-editing.png)
403
404 For a practical cleanup checklist, see the [Memory Guide](memory.md).
405
406 ## Backup And Restore
407
408 Backups protect your chats, projects, knowledge, memory, settings, skills, and
409 workspace files.
410
411 Create a backup before:
412
413 - major updates;
414 - plugin experiments;
415 - bulk memory cleanup;
416 - moving to a new machine;
417 - deleting or reorganizing important project files.
418
419 Open **Settings -> Backup & Restore** to create or restore a backup.
420
421 Secrets are sensitive and may not always be included in backup archives. Keep a
422 separate secure copy of credentials you depend on.
423
424 ## Next Steps
425
426 - [Quick Start](../quickstart.md)
427 - [Projects guide](projects.md)
428 - [Browser guide](browser.md)
429 - [A0 CLI Connector](a0-cli-connector.md)
430 - [Skills guide](skills.md)
431 - [Agent Profiles guide](agent-profiles.md)
432 - [Model Presets guide](model-presets.md)
433 - [MCP Setup](mcp-setup.md)
434 - [Troubleshooting](troubleshooting.md)
435 - [DeepWiki for Agent Zero](https://deepwiki.com/agent0ai/agent-zero)