OpenCode Advisor MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| OPENCODE_ADVISOR_TIMEOUT_MS | No | Timeout in milliseconds for the advisor request (default 120000). | |
| OPENCODE_ADVISOR_ALLOWED_ROOTS | Yes | Semicolon-separated list of allowed root directories for the advisor to access. | |
| OPENCODE_ADVISOR_MAX_DIFF_CHARS | No | Maximum number of characters in the diff sent to the advisor (default 60000). |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| ask_opencode_advisorC | Ask the local read-only OpenCode codex-advisor agent to review current git changes. |
| ask_opencode_plannerC | Ask the local read-only OpenCode planning partner to improve a plan without taking over implementation. |
| get_opencode_taskB | Check whether a queued or running OpenCode planner/reviewer task has finished. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 3 tools
Each tool has a clearly distinct purpose: ask_opencode_advisor reviews git changes, ask_opencode_planner improves plans, and get_opencode_task checks task status. No overlap.
All tool names follow a consistent verb_noun pattern in snake_case: ask_opencode_advisor, ask_opencode_planner, get_opencode_task.
With only 3 tools, the server is minimal but acceptable for a focused local agent interaction. However, it borders on too few for a complete agent workflow.
The server provides tools to ask agents and check task status, but lacks tools to retrieve results, cancel tasks, or configure agents, creating potential dead ends.