Fix docker agent serve xxx commands

Signed-off-by: Guillaume Tardif <guillaume.tardif@gmail.com>

Author: gtardif

content/manuals/ai/docker-agent/integrations/a2a.md

@@ -29,7 +29,7 @@ Before starting an A2A server, you need:
 Basic usage:
 
 ```console
-$ docker agent a2a ./agent.yaml
+$ docker agent serve a2a ./agent.yaml
 ```
 
 Your agent is now accessible via HTTP. Other A2A systems can discover your
@@ -38,19 +38,19 @@ agent's capabilities and call it.
 Custom port:
 
 ```console
-$ docker agent a2a ./agent.yaml --port 8080
+$ docker agent serve a2a ./agent.yaml --port 8080
 ```
 
 Specific agent in a team:
 
 ```console
-$ docker agent a2a ./agent.yaml --agent engineer
+$ docker agent serve a2a ./agent.yaml --agent engineer
 ```
 
 From OCI registry:
 
 ```console
-$ docker agent a2a agentcatalog/pirate --port 9000
+$ docker agent serve a2a agentcatalog/pirate --port 9000
 ```
 
 ## HTTP endpoints
@@ -142,7 +142,7 @@ Here's a concrete scenario where A2A is useful. You have two agents:
 Run the specialist as an A2A server:
 
 ```console
-$ docker agent a2a ./code-reviewer.yaml --port 8080
+$ docker agent serve a2a ./code-reviewer.yaml --port 8080
 Listening on 127.0.0.1:8080
 ```
 
@@ -198,7 +198,7 @@ agents:
 
 ## What's next
 
-- Review the [CLI reference](../reference/cli.md#a2a) for all `docker agent a2a`
+- Review the [CLI reference](../reference/cli.md#a2a) for all `docker agent serve a2a`
   options
 - Learn about [MCP mode](./mcp.md) to expose agents as tools in MCP clients
 - Learn about [ACP mode](./acp.md) for editor integration

content/manuals/ai/docker-agent/integrations/acp.md

@@ -131,11 +131,11 @@ reference:
 
 ```console
 # Local file path
-$ docker agent acp ./agent.yml
+$ docker agent serve acp ./agent.yml
 
 # OCI registry reference
-$ docker agent acp agentcatalog/pirate
-$ docker agent acp dockereng/myagent:v1.0.0
+$ docker agent serve acp agentcatalog/pirate
+$ docker agent serve acp dockereng/myagent:v1.0.0
 ```
 
 Use the same syntax in your editor configuration:
@@ -177,7 +177,7 @@ If the agent starts but can't access files or perform other actions, check:
 - Explore the [toolsets reference](../reference/toolsets.md) to learn what tools
   are available
 - Add [RAG for codebase search](../rag.md) to your agent
-- Check the [CLI reference](../reference/cli.md) for all `docker agent acp` options
+- Check the [CLI reference](../reference/cli.md) for all `docker agent serve acp` options
 - Browse [example
   configurations](https://github.com/docker/cagent/tree/main/examples) for
   inspiration

content/manuals/ai/docker-agent/integrations/mcp.md

@@ -93,8 +93,10 @@ Example configuration:
 {
   "mcpServers": {
     "myagent": {
-      "command": "/usr/local/bin/cagent",
+      "command": "/usr/local/bin/docker",
       "args": [
+        "agent", 
+        "serve",
         "mcp",
         "/path/to/agent.yml",
         "--working-dir",
@@ -111,7 +113,7 @@ Example configuration:
 
 Configuration breakdown:
 
-- `command`: Full path to your `cagent` binary (use `which cagent` to find it)
+- `command`: Full path to your `docker` binary (use `which docker` to find it), or path to docker-agent if not using the docker CLI plugin
 - `args`: MCP command arguments:
   - `mcp`: The subcommand to run cagent in MCP mode
   - `dockereng/myagent`: Your agent configuration (local file path or OCI
@@ -142,7 +144,7 @@ Command breakdown:
 - `myagent`: Name for this MCP server in Claude Code
 - `--env`: Pass environment variables (repeat for each variable)
 - `--`: Separates Claude Code options from the MCP server command
-- `docker agent mcp /path/to/agent.yml`: The cagent MCP command with the path to your
+- `docker agent serve mcp /path/to/agent.yml`: The cagent MCP command with the path to your
   agent configuration
 - `--working-dir $(pwd)`: Set the working directory for agent execution
 
@@ -153,7 +155,7 @@ sessions.
 
 For other MCP-compatible clients, you need to:
 
-1. Start cagent with `docker agent mcp /path/to/agent.yml --working-dir /project/path`
+1. Start cagent with `docker agent serve mcp /path/to/agent.yml --working-dir /project/path`
 2. Configure the client to communicate with Docker Agent over stdio
 3. Pass required environment variables (API keys, etc.)
 
@@ -166,11 +168,11 @@ reference:
 
 ```console
 # Local file path
-$ docker agent mcp ./agent.yml
+$ docker agent serve mcp ./agent.yml
 
 # OCI registry reference
-$ docker agent mcp agentcatalog/pirate
-$ docker agent mcp dockereng/myagent:v1.0.0
+$ docker agent serve mcp agentcatalog/pirate
+$ docker agent serve mcp dockereng/myagent:v1.0.0
 ```
 
 Use the same syntax in MCP client configurations:
@@ -179,8 +181,8 @@ Use the same syntax in MCP client configurations:
 {
   "mcpServers": {
     "myagent": {
-      "command": "/usr/local/bin/cagent",
-      "args": ["mcp", "agentcatalog/pirate"]
+      "command": "/usr/local/bin/docker",
+      "args": ["agent", "serve", "mcp", "agentcatalog/pirate"]
     }
   }
 }
@@ -299,7 +301,7 @@ around.
 - Explore the [toolsets reference](../reference/toolsets.md) to learn what tools
   agents can use
 - Add [RAG for codebase search](../rag.md) to your agent
-- Check the [CLI reference](../reference/cli.md) for all `docker agent mcp` options
+- Check the [CLI reference](../reference/cli.md) for all `docker agent serve mcp` options
 - Browse [example
   configurations](https://github.com/docker/cagent/tree/main/examples) for
   different agent types

content/manuals/ai/docker-agent/reference/cli.md

@@ -53,7 +53,7 @@ systems to discover and interact with your agent. Auto-selects an available
 port if not specified.
 
 ```console
-$ docker agent a2a agent-file|registry-ref
+$ docker agent serve a2a agent-file|registry-ref
 ```
 
 > [!NOTE]
@@ -77,8 +77,8 @@ Supports [runtime flags](#runtime-flags).
 Examples:
 
 ```console
-$ docker agent a2a ./agent.yaml --port 8080
-$ docker agent a2a agentcatalog/pirate --port 9000
+$ docker agent serve a2a ./agent.yaml --port 8080
+$ docker agent serve a2a agentcatalog/pirate --port 9000
 ```
 
 ### acp
@@ -87,7 +87,7 @@ Start agent as ACP (Agent Client Protocol) server on stdio for editor integratio
 See [ACP integration](../integrations/acp.md) for setup guides.
 
 ```console
-$ docker agent acp agent-file|registry-ref
+$ docker agent serve acp agent-file|registry-ref
 ```
 
 Arguments:
@@ -146,7 +146,7 @@ Arguments:
 HTTP API server.
 
 ```console
-$ docker agent api agent-file|agents-dir
+$ docker agent serve api agent-file|agents-dir
 ```
 
 Arguments:
@@ -166,9 +166,9 @@ Supports [runtime flags](#runtime-flags).
 Examples:
 
 ```console
-$ docker agent api ./agent.yaml
-$ docker agent api ./agents/ --listen :9000
-$ docker agent api docker.io/user/agent --pull-interval 10
+$ docker agent serve api ./agent.yaml
+$ docker agent serve api ./agents/ --listen :9000
+$ docker agent serve api docker.io/user/agent --pull-interval 10
 ```
 
 The `--pull-interval` flag works only with OCI references. Automatically pulls and reloads at the specified interval.
@@ -304,7 +304,7 @@ MCP (Model Context Protocol) server on stdio. Exposes agents as tools to MCP
 clients. See [MCP integration](../integrations/mcp.md) for setup guides.
 
 ```console
-$ docker agent mcp agent-file|registry-ref
+$ docker agent serve mcp agent-file|registry-ref
 ```
 
 Arguments:
@@ -316,8 +316,8 @@ Supports [runtime flags](#runtime-flags).
 Examples:
 
 ```console
-$ docker agent mcp ./agent.yaml
-$ docker agent mcp docker.io/user/agent:latest
+$ docker agent serve mcp ./agent.yaml
+$ docker agent serve mcp docker.io/user/agent:latest
 ```
 
 ### new

content/manuals/ai/docker-agent/sharing-agents.md

@@ -82,8 +82,8 @@ Agents can be exposed as tools in MCP clients:
 {
   "mcpServers": {
     "myagent": {
-      "command": "/usr/local/bin/cagent",
-      "args": ["mcp", "agentcatalog/pirate"]
+      "command": "/usr/local/bin/docker",
+      "args": ["agent", "serve", "mcp", "agentcatalog/pirate"]
     }
   }
 }