You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
<p>macOS and Linux are fully supported for development. On Windows, use <code>task build-local</code> to build via Docker.</p>
@@ -86,7 +86,7 @@ Key conventions:
86
86
87
87
File issues on the [GitHub issue tracker](https://github.com/docker/docker-agent/issues). Please:
88
88
89
-
<divclass="callout callout-info">
89
+
<divclass="callout callout-info"markdown="1">
90
90
<divclass="callout-title">ℹ️ See also
91
91
</div>
92
92
<ahref="{{ '/community/troubleshooting/' | relative_url }}">Troubleshooting</a> — Common issues and debug mode. <ahref="{{ '/community/telemetry/' | relative_url }}">Telemetry</a> — What data is collected and how to opt out.
@@ -105,7 +105,7 @@ File issues on the [GitHub issue tracker](https://github.com/docker/docker-agent
105
105
4.**Sign** your commits with `git commit -s` (DCO required)
106
106
5.**Open a pull request** against the `main` branch
107
107
108
-
<divclass="callout callout-tip">
108
+
<divclass="callout callout-tip"markdown="1">
109
109
<divclass="callout-title">💡 Tip
110
110
</div>
111
111
<p>Use the dogfooding agent (<code>docker agent run ./golang_developer.yaml</code>) to help write and review your changes before submitting.</p>
<p>Always enable <code>--debug</code> when reporting issues. The log file contains detailed traces of API calls, tool executions, and agent interactions.</p>
@@ -146,7 +146,7 @@ docker-agent validates config at startup and reports errors with line numbers. C
146
146
- MCP toolsets need either `command` (stdio), `remote` (SSE/HTTP), or `ref` (Docker)
147
147
- Provider names must be one of: `openai`, `anthropic`, `google`, `amazon-bedrock`, `dmr`, etc.
148
148
149
-
<divclass="callout callout-info">
149
+
<divclass="callout callout-info"markdown="1">
150
150
<divclass="callout-title">ℹ️ Schema Validation
151
151
</div>
152
152
<p>Use the <ahref="https://github.com/docker/docker-agent/blob/main/agent-schema.json">JSON schema</a> in your editor for real-time config validation and autocompletion.</p>
@@ -234,7 +234,7 @@ When reviewing debug logs, search for these key patterns:
234
234
|`[RAG Manager]`| RAG retrieval operations |
235
235
|`[Reranker]`| Reranking operations |
236
236
237
-
<divclass="callout callout-warning">
237
+
<divclass="callout callout-warning"markdown="1">
238
238
<divclass="callout-title">⚠️ Still stuck?
239
239
</div>
240
240
<p>If these steps don't resolve your issue, file a bug on the <ahref="https://github.com/docker/docker-agent/issues">GitHub issue tracker</a> with your debug log attached, or ask on <ahref="https://dockercommunity.slack.com/archives/C09DASHHRU4">Slack</a>.</p>
Copy file name to clipboardExpand all lines: docs/concepts/agents/index.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -36,7 +36,7 @@ agents:
36
36
37
37
Every docker-agent configuration has a **root agent** — the entry point that receives user messages. In a single-agent setup, this is the only agent. In a multi-agent setup, the root agent acts as a coordinator, delegating tasks to specialized sub-agents.
38
38
39
-
<div class="callout callout-info">
39
+
<div class="callout callout-info" markdown="1">
40
40
<div class="callout-title">ℹ️ Naming
41
41
</div>
42
42
<p>The first agent defined in your YAML (or the one named <code>root</code>) is the root agent by default. You can also specify which agent to start with using <code>docker agent run config.yaml -a agent_name</code>.</p>
Copy file name to clipboardExpand all lines: docs/concepts/distribution/index.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -12,7 +12,7 @@ _Package, share, and run agents via OCI-compatible registries — just like cont
12
12
13
13
docker-agent agents can be pushed to any OCI-compatible registry (Docker Hub, GitHub Container Registry, etc.) and pulled/run anywhere. This makes sharing agents as easy as sharing Docker images.
14
14
15
-
<divclass="callout callout-tip">
15
+
<divclass="callout callout-tip"markdown="1">
16
16
<divclass="callout-title">💡 Tip
17
17
</div>
18
18
<p>For CLI commands related to distribution, see <ahref="{{ '/features/cli/' | relative_url }}">CLI Reference</a> (<code>docker agent share push</code>, <code>docker agent share pull</code>, <code>docker agent alias</code>).</p>
Copy file name to clipboardExpand all lines: docs/concepts/models/index.md
+1-1Lines changed: 1 addition & 1 deletion
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -97,7 +97,7 @@ models:
97
97
thinking_budget: none # disable thinking
98
98
```
99
99
100
-
<div class="callout callout-info">
100
+
<div class="callout callout-info" markdown="1">
101
101
<div class="callout-title">ℹ️ Multi-provider teams
102
102
</div>
103
103
<p>Different agents can use different providers in the same config. See <a href="{{ '/concepts/multi-agent/' | relative_url }}">Multi-Agent</a> for patterns.</p>
Copy file name to clipboardExpand all lines: docs/concepts/multi-agent/index.md
+6-6Lines changed: 6 additions & 6 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -34,7 +34,7 @@ docker-agent supports two multi-agent patterns:
34
34
35
35
You can combine both patterns in the same configuration — an agent can have both `sub_agents` and `handoffs`.
36
36
37
-
<divclass="callout callout-tip">
37
+
<divclass="callout callout-tip"markdown="1">
38
38
<divclass="callout-title">💡 When to use which
39
39
</div>
40
40
<p><strong><code>sub_agents</code></strong> — Use when a coordinator needs to send tasks to specialists and synthesize their results.</p>
@@ -62,7 +62,7 @@ transfer_task(
62
62
)
63
63
```
64
64
65
-
<divclass="callout callout-info">
65
+
<divclass="callout callout-info"markdown="1">
66
66
<divclass="callout-title">ℹ️ Auto-Approved
67
67
</div>
68
68
<p>Unlike other tools, <code>transfer_task</code> is always auto-approved — no user confirmation needed. This allows seamless delegation between agents.</p>
<p>Each agent can only hand off to agents listed in its own <code>handoffs</code> array. The <code>handoff</code> tool is automatically injected — you don't need to add it manually.</p>
@@ -140,7 +140,7 @@ agents:
140
140
- root
141
141
```
142
142
143
-
<div class="callout callout-tip">
143
+
<div class="callout callout-tip" markdown="1">
144
144
<div class="callout-title">💡 Full pipeline example
145
145
</div>
146
146
<p>For a more complex handoff graph with branching and multiple processing stages, see <a href="https://github.com/docker/docker-agent/blob/main/examples/handoff.yaml"><code>examples/handoff.yaml</code></a>.</p>
@@ -212,7 +212,7 @@ External sub-agents are automatically named after their last path segment — fo
212
212
- reviewer:docker.io/myorg/review-agent:latest
213
213
```
214
214
215
-
<div class="callout callout-tip">
215
+
<div class="callout callout-tip" markdown="1">
216
216
<div class="callout-title">💡 Tip
217
217
</div>
218
218
<p>External sub-agents work with any OCI-compatible registry, not just the Docker Agent Catalog. See <a href="{{ '/concepts/distribution/' | relative_url }}">Agent Distribution</a> for more on registry references.</p>
@@ -345,7 +345,7 @@ toolsets:
345
345
- **Use the right model** — Use capable models for complex reasoning, cheap models for simple tasks
346
346
- **Choose the right pattern** — Use `sub_agents` for hierarchical task delegation, `handoffs` for pipeline workflows and conversational routing
347
347
348
-
<div class="callout callout-info">
348
+
<div class="callout callout-info" markdown="1">
349
349
<div class="callout-title">ℹ️ Beyond docker-agent
350
350
</div>
351
351
<p>For interoperability with other agent frameworks, docker-agent supports the <a href="{{ '/features/a2a/' | relative_url }}">A2A protocol</a> and can expose agents via <a href="{{ '/features/mcp-mode/' | relative_url }}">MCP Mode</a>.</p>
Copy file name to clipboardExpand all lines: docs/concepts/tools/index.md
+2-2Lines changed: 2 additions & 2 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -17,7 +17,7 @@ When an agent needs to perform an action, it makes a **tool call**. The docker-a
17
17
3. docker-agent executes the tool and returns the result
18
18
4. Agent incorporates the result and responds
19
19
20
-
<divclass="callout callout-info">
20
+
<divclass="callout callout-info"markdown="1">
21
21
<divclass="callout-title">ℹ️ Tool Confirmation
22
22
</div>
23
23
<p>By default, docker-agent asks for user confirmation before executing tools that have side effects (shell commands, file writes). Use <code>--yolo</code> to auto-approve all tool calls.</p>
@@ -60,7 +60,7 @@ toolsets:
60
60
61
61
See [Tool Config]({{ '/configuration/tools/#mcp-tools' | relative_url }}) for full MCP configuration reference.
62
62
63
-
<div class="callout callout-tip">
63
+
<div class="callout callout-tip" markdown="1">
64
64
<div class="callout-title">💡 See also
65
65
</div>
66
66
<p>For full configuration reference, see <a href="{{ '/configuration/tools/' | relative_url }}">Tool Config</a>. For RAG (document retrieval), see <a href="{{ '/features/rag/' | relative_url }}">RAG</a>.</p>
<p>Default is <code>0</code> (unlimited). Always set <code>max_iterations</code> for agents with powerful tools like <code>shell</code> to prevent infinite loops. A value of 20–50 is typical for development agents.</p>
Copy file name to clipboardExpand all lines: docs/configuration/hooks/index.md
+4-4Lines changed: 4 additions & 4 deletions
Display the source diff
Display the rich diff
Original file line number
Diff line number
Diff line change
@@ -12,7 +12,7 @@ _Run shell commands at various points during agent execution for deterministic c
12
12
13
13
Hooks allow you to execute shell commands or scripts at key points in an agent's lifecycle. They provide deterministic control that works alongside the LLM's behavior, enabling validation, logging, environment setup, and more.
<p>Hooks run synchronously and can slow down agent execution. Keep hook scripts fast and efficient. Consider using <code>suppress_output: true</code> for logging hooks to reduce noise.</p>
221
221
222
222
</div>
223
223
224
-
<div class="callout callout-info">
224
+
<div class="callout callout-info" markdown="1">
225
225
<div class="callout-title">ℹ️ Session End and Cancellation
226
226
</div>
227
227
<p><code>session_end</code> hooks are designed to run even when the session is interrupted (e.g., Ctrl+C). They are still subject to their configured timeout.</p>
@@ -376,7 +376,7 @@ $ docker agent run agentcatalog/coder \
376
376
--hook-pre-tool-use "./audit.sh"
377
377
```
378
378
379
-
<div class="callout callout-info">
379
+
<div class="callout callout-info" markdown="1">
380
380
<div class="callout-title">ℹ️ Merging behavior
381
381
</div>
382
382
<p>CLI hooks are <strong>appended</strong> to any hooks already defined in the agent's YAML config. They don't replace existing hooks. Pre/post-tool-use hooks added via CLI match all tools (equivalent to <code>matcher: "*"</code>).</p>
0 commit comments