Shellway

Start an advanced search

start_search

Read-only

Search a folder in the background by file name or by text inside files, with optional regular expression, file-type filter (for example ".js,.ts") and lines of context around each match. Returns the first results and a search_id; use get_more_search_results for the rest. Good for big folders and many matches.

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": {
      "path": "<path>",
      "pattern": "<pattern>"
    },
    "name": "start_search"
  }
}

Input Schema

Name	Required	Description
`mode`	No	Default "name".
`path`	Yes	Folder to search in.
`regex`	No	Treat pattern as a regular expression. Default false.
`device`	No	Device name. Optional when the account has only one online device.
`pattern`	Yes	What to look for: a name pattern (name mode) or text (content mode). Treated as a regular expression when regex is true.
`ignore_case`	No	Default true.
`max_results`	No	Results to return per call, default 100, max 500.
`file_pattern`	No	Content mode only: only search files whose name matches, for example ".js" or ".js,*.ts".
`context_lines`	No	Content mode only: lines to show before and after each match, 0 to 5. Default 0.

Schema Changelog

Changes observed during successful MCP inspections.

First observedOct 11, 2026

TDQS

A3.9/5.0

Behavior4/5

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

Annotations already establish the safe read-only profile, so the description's job here is lighter. It still adds real operational context: the search runs in the background, it returns only the first results plus a search_id, and continuation is via get_more_search_results. The session lifecycle (e.g. cancelling via stop_search) is left undisclosed.

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?

Purpose and scope are front-loaded in the first clause, and the second sentence carries the two facts an agent most needs (return shape and the continuation tool). The single packed sentence is dense but each phrase maps to a real capability, with little wasted wording.

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

Completeness4/5

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

With no output schema, the description correctly compensates by explaining that the call returns first results plus a search_id. For a 9-parameter tool the prose plus full schema coverage is largely sufficient, leaving only the stop/session-management aspect uncovered.

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 100%, so every parameter is already documented, including the '*.js,*.ts' and context_lines examples the description repeats. The description restates mode/regex/file-pattern concepts without adding syntax or defaults beyond the schema, so the baseline of 3 is appropriate.

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

Purpose4/5

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

States a specific verb and resource ('search a folder ... by file name or by text inside files') with concrete modifiers (regex, file-type filter, context lines). The 'in the background' framing implicitly separates it from the synchronous search_files sibling, but that sibling is never named, so differentiation is inferred rather than stated.

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?

The closing line 'Good for big folders and many matches' gives a clear when-to-use condition, and it routes the agent to get_more_search_results for the remainder. It stops short of explicit exclusions (when a plain/synchronous search is preferable) and never mentions stop_search for cleanup.

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

All Shellway tools Inspect

Try in Browser

Claude Code, inside the Godot 4 editorFlockbay puts an AI agent in the real Godot 4 editor: it writes GDScript, runs your game and fixes what breaks. Free, or bring your own Claude Code or Codex.flockbay.comAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?