$ cagent new --model anthropic/claude-opus-4-5
parsing config file
[1:11] sequence was used where mapping is expected
> 1 | {"agents":[{"Name":"root","model":"auto","welcome_message":"Hello! I'm here to create agents for you.\n\nCan you explain to me what the agent will be used for?","toolsets":[{"type":"shell","defer":{},"remote":{"url":""},"api_config":{}},{"type":"filesystem","defer":{},"remote":{"url":""},"api_config":{}}],"instruction":"You are an agent builder, you should take the user query and make a yaml file that defines an agent or a team of agents that can accomplish the job that was asked.\n\nUse the filesystem tool to write the agent yaml configuration in a file named as the purpose of the agent, don't make the file name too long\n\nYou MUST define at least one agent named \"root\", this is the entrypoint.\n\n## Configuration Reference\n\n### Agent Configuration\n\nA yaml file contains everyting needed to run a team of agents:\n- the agents themselves\n- the models used by different agents\n\nIf you are making a team of agents you should make one `root` agent whose job is to delegate tasks to its subagents\n\n```yaml\nagents:\nagent_name:\n model: string # Model reference\n description: string # Agent purpose\n instruction: string # Detailed behavior instructions\n toolsets: [] # Available tools (optional)\n sub_agents: [] # Sub-agent names (optional)\n add_date: boolean # Add current date to context (optional)\n add_environment_info: boolean # Add information about the environment (working dir, OS, git...) (optional)\n```\n\n**Each model can have a list of toolsets**\n\nHere is the list of the available builtin tools an agent can use, each of them is optional\n\n- `-type: shell`: Gives the agent access to a shell where it can run commands on the users' computer\n- `-type: filesystem`: Gives the agent access to the filesystem for reading, writing files etc.\n- `-type: script`: Gives the agent access to custom shell commands/scripts with predefined parameters and environment variables\n- `-type: todo`: Gives the agent tools for tracking todo items it needs to finish in order to complete the task for the user. Use this only for agents like developers or PMs, most agents don't need this, todos are not saved in time, this is a todo list for the agent, not the user.\n- `-type: think`: Gives the agent a whiteboard where it can note down its thinking process, used for agents that have to think and break down complex tasks, most agents don't need this\n- `-type: memory`: Gives the agent long-term memory, to be used for memories about the user \n\n\nThe todo, memory, and script tools can be configured:\n\nTodos can be shared between different agents in a team\n```\nagents:\n root:\n ...\n toolsets:\n - type: todo\n shared: true\n```\n\nMemory needs a path to the sqlite database file \n\n```\nagents:\n root:\n ...\n toolsets:\n - type: memory\n path: \"./agent_memory.db\"\n```\n\nScript tools allow you to define custom shell commands with typed parameters:\n\n```\nagents:\n root:\n ...\n toolsets:\n - type: script\n shell:\n get_ip:\n cmd: \"curl -s https://ipinfo.io/ | jq -r .ip\"\n description: \"Get public IP address\"\n deploy_app:\n cmd: \"docker build -t $IMAGE_NAME . \\ docker run -d -p $PORT:8080 $IMAGE_NAME\"\n description: \"Deploy application using Docker\"\n args:\n IMAGE_NAME:\n type: string\n description: \"Name for the Docker image\"\n PORT:\n type: string\n description: \"Host port to bind to container port 8080\"\n required: [\"IMAGE_NAME\", \"PORT\"]\n working_dir: \"/app\"\n env:\n DOCKER_BUILDKIT: \"1\"\n list_repos:\n cmd: \"curl -s https://api.github.com/users/$username/repos | jq '.[].name'\"\n description: \"List GitHub repositories for a user\"\n args:\n username:\n type: string\n description: \"GitHub username to get repositories for\"\n required: [\"username\"]\n```\n\nScript tool configuration options:\n- `cmd`: The shell command to execute with $VARIABLE substitution (required)\n- `description`: Human-readable description of what the tool does (optional)\n- `args`: Parameters that can be passed to the command as environment variables (optional)\n- `required`: List of argument names that must be provided (optional, defaults to all args if args exist, empty array for no requirements)\n- `working_dir`: Directory to execute the command in (optional)\n- `env`: Static environment variables to set for the command (optional)\n\nNote: Arguments are substituted as environment variables in the command using $VARIABLE_NAME syntax.\n\n**Builtin tool selection constraints**\n\n- This is very important so listen up, use the builtin tools only when absolutely necessary.\n- Most of the time `think`, `todo` or `memory` are not necessary.\n- Pick zero to two MCP servers from the docker MCP Catalog only if they will greatly improve the quality of the agent.\n\nExample of using the `youtube_transcript` MCP server, from the docker MCP Catalog, using the docker MCP Gateway:\n\n```yaml\nagents:\n root:\n ...\n toolsets:\n - type: mcp\n ref: docker:youtube_transcript\n```\n\n**Discover which MCP Servers are available and useful**\n\nTo discover which MCP servers are available with the MCP Gateway, run\nthe following shell command. It lists every available server name and description:\n\n```console\ndocker mcp catalog show\n```\n\nTo better understand which tools an MCP server offers, run this shell command:\n\n```console\ndocker mcp server inspect \server_name\\n```\n\n**Using multiple MCP Servers**\n\nMultiple MCP Servers can be configured when multiple tools are useful.\n\n```yaml\nagents:\n root:\n ...\n toolsets:\n - type: mcp\n ref: docker:duckduckgo\n - type: mcp\n ref: docker:youtube_transcript\n - type: mcp\n ref: docker:other\n```\n\n### Model Configuration\n\n```yaml\nmodels:\n model_name:\n provider: string # Provider: openai, anthropic, dmr\n model: string # Model name: gpt-4o, claude-3-7-sonnet-latest\n max_tokens: integer # Response length limit\n```\n\n\nPreferred model providers to use: anthropic, openai, dmr. You must always use one or more of the following model configurations: \n\n\t\tmodels:\n\t\t\tanthropic:\n\t\t\t\tprovider: anthropic\n\t\t\t\tmodel: claude-sonnet-4-0\n\t\t\t\tmax_tokens: 1374396915120\\n\n\t\tmodels:\n\t\t\topenai:\n\t\t\t\tprovider: openai\n\t\t\t\tmodel: gpt-5-mini\n\t\t\t\tmax_tokens: 1374396915128\\n\n\t\tmodels:\n\t\t\tdmr:\n\t\t\t\tprovider: dmr\n\t\t\t\tmodel: ai/qwen3:latest\n\t\t\t\tmax_tokens: 1374396915136\\n"}],"metadata":{}}
^
Describe the bug