scope_task
Given a task description and repo path, returns the minimal set of files to read, utilities to reuse, downstream callers at risk, and an overall risk level.
DESCRIPTION
This is the first tool to call before any non-trivial coding task. It prevents agents from reading half the repo and helps them start in the right files.
WHEN TO USE
GOOD FIT
- At the start of every non-trivial coding task.
- When you do not know which files a change touches.
- When you want to avoid unnecessary repo reads.
- When you need to know whether reusable utilities already exist.
AVOID IT WHEN
- You already know the exact files involved.
- You are only answering a conceptual question, not planning code work.
- You need a specific symbol implementation now; use read_code.
PARAMETERS
| Name | Type | Required | Description |
|---|---|---|---|
| path | string | Yes | Absolute path to the repo root. |
| task | string | Yes | Plain-English description of the coding task. |
EXAMPLE
tools-call.json
{ "jsonrpc": "2.0", "id": 9, "method": "tools/call", "params": { "name": "scope_task", "arguments": { "path": "/workspace/repo", "task": "Add billing FAQ and rate limit docs pages" } }}OUTPUT
response.json
{ "jsonrpc": "2.0", "id": 9, "result": { "content": [ { "type": "text", "text": "Relevant files (max 7): src/app/docs/layout.tsx, src/components/docs/DocsSidebar.tsx, src/components/docs/docs-content.ts\nReuse: PageHeader, CodeBlock, SectionDivider\nDownstream risk: sidebar ordering, TOC mapping\nRisk: medium" } ] }}TOKEN EFFICIENCY
This tool saves the most tokens when it prevents unnecessary exploration. Its value is in reducing total context consumption across the whole task.