cc-delegate
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| DELEGATE_MODEL | No | Model identifier for the delegate worker (default: MiniMax M3) | |
| DELEGATE_API_KEY | Yes | API key for the delegate worker's Anthropic-compatible endpoint | |
| DELEGATE_BASE_URL | No | Base URL for the delegate worker's Anthropic-compatible endpoint (default varies) | |
| DELEGATE_MAX_TURNS | No | Maximum number of turns for the delegate worker | |
| DELEGATE_TIMEOUT_MS | No | Timeout in milliseconds for the delegate worker | |
| DELEGATE_MAX_BUDGET_USD | No | Maximum budget in USD for the delegate worker |
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 |
|---|---|
| run_dev_taskA | Starts an autonomous coding worker (on whatever model DELEGATE_MODEL points to) on an isolated git worktree. Returns a task_id immediately; poll with get_task_status, then fetch_task_result. |
| get_task_statusC | Returns current status, progress, cost and turns. |
| fetch_task_resultB | When status is 'succeeded', returns summary, patch path, files changed, tests and cost. |
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 clear, distinct purpose: run_dev_task initiates a task, get_task_status checks progress, and fetch_task_result retrieves the final output. No overlap.
All tool names follow a consistent verb_noun pattern in snake_case (run_dev_task, get_task_status, fetch_task_result), making them predictable.
Three tools is exactly right for a focused task life cycle: start, poll, and fetch result. Not too few or too many.
The tool set covers the full life cycle of an autonomous coding task: creation, status polling, and result retrieval. No obvious gaps for the stated purpose.