agentleFS
D-SheyD-Shey/autoharness/.agents/skills/codemode-types/SKILL.md

codemode-types

Skill10 starsChanged todayMITView on GitHub

TypeScript definition generation from MCP Tool JSON Schema

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 3 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈3

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use when compiling MCP server tool definitions and JSON Schemas into compact, strongly-typed TypeScript definitions:

    line 9
  2. Distinctive:

    Run unit tests against representative MCP schemas (nested objects, arrays, enums, optional fields).

    line 40
  3. Distinctive:

    Verify JSDoc descriptions match the tool descriptions.

    line 42

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

codemode-types

When to use

Use when compiling MCP server tool definitions and JSON Schemas into compact, strongly-typed TypeScript definitions:

  • Providing code-generating models with type signatures instead of verbose JSON schemas.
  • Exposing registered MCP tools as typed methods in sandbox scripts.
  • Filtering and searching tools by name or description with harness_search.

Steps

  1. Extract MCP tool definitions:
    • For each active server in Gateway, retrieve the list of tools and their inputSchema.
  2. Convert JSON Schema to TypeScript types:
    • Map primitive types: string -> string, number/integer -> number, boolean -> boolean, null -> null.
    • Map complex types: array -> T[], object -> { [prop: string]: T } or structured interfaces with optional (?) markers based on required arrays.
    • Map enum types to union literals: 'val1' | 'val2'.
    • Preserve description fields as JSDoc comments preceding properties and methods.
  3. Assemble namespaces and exports:
    • Wrap each server's tools into a typed namespace or interface:
      export interface ServerTools {
        tool_name(args: ToolArgs): Promise<unknown>;
      }
    • Export top-level global mcp object declaration:
      declare const mcp: {
        [serverName: string]: ServerTools;
      };
  4. Search and compaction:
    • Implement search(query) over tool names, descriptions, and server names.
    • Return only matched .d.ts blocks to minimize context token usage.

How to verify

  1. Run unit tests against representative MCP schemas (nested objects, arrays, enums, optional fields).
  2. Validate generated TypeScript definitions with a TypeScript compiler or syntax parser.
  3. Verify JSDoc descriptions match the tool descriptions.

Facts found here

  • MCP schemas strictly adhere to JSON Schema draft-07.
  • Input schemas always specify type: "object" at the root for tool argument dictionaries.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.