opencode_config_set

Patch the OpenCode config on the local or a remote Yaver machine. Each top-level field is optional; pass empty string to clear. defaultAgent picks the agent invoked when no --agent flag is given. model is the top-level default model id. smallModel is used for cheap helper calls. buildModel and pl...

Server Yaver yaver-cli
Category Write
Risk class Medium
Parameters 70 required

What opencode_config_set does on Yaver

AI agents use opencode_config_set to create or update resources in Yaver — usually the action step of a workflow, after the agent has gathered context. Every call changes real data in your Yaver environment.

ParameterTypeRequiredDescription
model string Top-level default model id
device_id string Optional remote device ID
providers array Optional list of provider upserts — each {id, name?, baseUrl?, apiKey?, models?, delete?}.
plan_model string Model for agent.plan
build_model string Model for agent.build
small_model string Cheap-helper model id
default_agent string Default agent name (e.g. build, plan, or any custom agent)

Parameters from the server's own tool schema.

Why opencode_config_set needs a policy

An AI agent can call opencode_config_set faster than any human can review — one bad instruction and it creates or modifies resources in Yaver by the hundred, each call as confident as the last.

Risk signalsHandles credentials or secrets (providers[].apiKey) · High parameter count (13 properties)

Questions about opencode_config_set

What does the opencode_config_set tool do? +

Patch the OpenCode config on the local or a remote Yaver machine. Each top-level field is optional; pass empty string to clear. defaultAgent picks the agent invoked when no --agent flag is given. model is the top-level default model id. smallModel is used for cheap helper calls. buildModel and planModel set the model under agent.build / agent.plan in opencode.json. providers is an optional list of provider upserts — each entry creates or merges a provider entry by id. Common cases: BYOK OpenRouter via {id:'openrouter', baseUrl:'https://openrouter.ai/api/v1', apiKey:'...'} or pointing a remote machine's opencode at a Tailscale-reachable Ollama via {id:'ollama', baseUrl:'http://100.x.x.x:11434'}. Pass delete:true on a provider entry to remove it. Other config keys (custom agents, MCP servers) stay untouched. Returns the new config summary. It is categorised as a Write tool in the Yaver MCP Server, which means it can create or modify data. Consider rate limits to prevent runaway writes.

What parameters does opencode_config_set accept? +

opencode_config_set accepts 7 parameters: model, device_id, providers, plan_model, build_model, small_model, default_agent. The full parameter table on this page comes from the server's own tool schema.

How do I enforce a policy on opencode_config_set? +

Register the Yaver MCP server in PolicyLayer and add a rule for opencode_config_set: allow, deny, rate-limit, or require approval. Point your MCP client at the PolicyLayer proxy URL and the rule is enforced on every call, before it reaches Yaver. Nothing to install.

What risk level is opencode_config_set? +

opencode_config_set is a Write tool with medium risk. Write tools should be rate-limited to prevent accidental bulk modifications.

Can I rate-limit opencode_config_set? +

Yes. Add a rate_limit block to the opencode_config_set rule in your PolicyLayer policy. For example, setting max: 10 and window: 60 limits the tool to 10 calls per minute. Rate limits are tracked per agent session and reset automatically.

How do I block opencode_config_set completely? +

Set action: deny in the PolicyLayer policy for opencode_config_set. The AI agent will receive a policy violation error and cannot call the tool. You can also include a reason field to explain why the tool is blocked.

What MCP server provides opencode_config_set? +

opencode_config_set is provided by the Yaver MCP server (yaver-cli). PolicyLayer sits as a proxy in front of this server to enforce policies before tool calls reach the server.

// GET IN TOUCH

Have a question or want to learn more? Send us a message.

Message sent.

We'll get back to you soon.