droid-mcp

droid_start

Start a new Droid session in an allowed directory and send the first prompt; model, reasoning, and mode settings are validated before anything runs.

Instructions

Start a new Droid session in cwd (must be inside the allowed roots) and send it the first prompt. Retrying with the same idempotencyKey returns the same session. Pick a model with droid_models: a built-in id (claude-opus-5-5), a BYOK custom id from ~/.factory/settings.json (custom::), a tier alias (anthropic-latest-premium) or a version wildcard (claude-opus-*). reasoningEffort must be one the model supports. mission needs autonomy 'high'. Unknown models and invalid combinations are refused before anything runs.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "cwd": "<cwd>",
      "prompt": "<prompt>",
      "idempotencyKey": "<idempotencyKey>"
    },
    "name": "droid_start"
  }
}

Input Schema

Name	Required	Description
`cwd`	Yes
`fork`	No	Fork this session id instead of starting fresh (start only)
`tags`	No	Tag names to add to the session
`model`	No	Model id from droid_models
`prompt`	Yes
`mission`	No	Switch to mission mode (multi-agent); needs autonomy "high"
`useSpec`	No	Switch to spec mode: plan before editing
`addTools`	No	Add these tool ids or selectors
`autonomy`	No
`worktree`	No	Start in a git worktree; a string names the branch (start only)
`onlyTools`	No	Use only these tool ids or MCP:<server>[/<tool>] selectors
`specModel`	No	Model for spec mode (defaults to model)
`removeTools`	No	Remove these tool ids or selectors
`workerModel`	No	Mission worker model (needs mission)
`idempotencyKey`	Yes
`validatorModel`	No	Mission validator model (needs mission)
`reasoningEffort`	No	Must be one the model supports
`appendSystemPrompt`	No	Text appended to the system prompt (start only)
`specReasoningEffort`	No
`workerReasoningEffort`	No
`validatorReasoningEffort`	No

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.0Nov 13, 2026

TDQS

A4/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries the full burden, and it does disclose key traits: validation happens before anything runs (unknown models and invalid combinations are refused), and retrying with the same idempotencyKey returns the same session rather than creating a duplicate. It does not describe what happens to an existing session on retry with a different payload, nor any permission model beyond the allowed-roots constraint, but the idempotency and pre-flight validation disclosures are substantial.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Compact and front-loaded: creation and cwd constraint first, then idempotency, then model selection, then validation. The model-id enumeration is dense but each form is distinct and useful. It is slightly list-like in the middle, but no sentence is filler.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

This is a 21-parameter mutation-style tool with no annotations and no output schema, so the bar is high. The description covers the required three parameters, the highest-risk validation rules, and model selection, but leaves the majority of optional parameters (tool selection, spec/mission sub-models, fork/worktree semantics) unexplained, and the return shape is unstated since there is no output schema to rely on.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 67%, and the description adds meaning for the highest-risk parameters: model id forms (built-in, custom:<provider>:<model>, tier alias, wildcard), reasoningEffort must be model-supported, and mission requires autonomy 'high'. However, 14 optional parameters (fork, worktree, tags, onlyTools/removeTools/addTools, spec/worker/validator model and reasoning fields, appendSystemPrompt, useSpec) get no explanation beyond short schema descriptions, and the cross-parameter dependency of workerModel/validatorModel on mission is only implied.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

States a specific verb+resource ("Start a new Droid session"), names the working directory and prompt as the initial payload, and distinguishes it from siblings by pointing to droid_models for model selection. An agent can tell this is the session-creation entry point versus droid_find, droid_send, or droid_result.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Gives clear preconditions (cwd must be inside the allowed roots) and routes model choice to droid_models, plus states mission needs autonomy 'high'. It does not explicitly say when to use fork/worktree versus a fresh start beyond the 'start only' tags, and gives no explicit when-not-to-use guidance such as preferring droid_respond for an existing session.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Deploy Server