Look up the exact JSON schema a specific AI coding agent uses for its edit tool.
Build a custom tool that matches a model's trained schema shape to avoid performance loss.
Browse collected tool schemas across providers using the included web interface.
Reference the pseudo-code implementations for string replacement and patch application edit tools.
| hacktronai/tools | 000madz000/payload-test-api-route-handler | akash-sr/memobsidian | |
|---|---|---|---|
| Stars | 3 | 3 | 3 |
| Language | TypeScript | TypeScript | TypeScript |
| Setup difficulty | easy | easy | moderate |
| Complexity | 2/5 | 2/5 | 3/5 |
| Audience | developer | developer | general |
Figures from each repo's GitHub metadata at analysis time.
This is a reference collection of JSON files rather than a runnable application.
HacktronAI/tools is a small reference repository that collects the tool schemas used by different AI coding agents. When a coding assistant like Claude Code or similar agents gives its underlying model access to tools, such as an edit tool for changing files, that tool has to be described to the model in a very specific shape, usually as a JSON schema listing the tool's name, what it does, and what parameters it expects. Different agent products define these tool shapes differently, even for the same underlying task. The README explains why this matters: if you hand a model a tool shaped differently than the one it was actually trained on, especially something as sensitive as an edit tool, you risk the model making mistakes or producing schema errors, which hurts its performance. This project exists to document those real, verbatim tool definitions from each agent so people can see exactly how each one is shaped, rather than guessing or reinventing them. The repository is organized by provider. Each provider gets its own folder containing a metadata file describing the provider, agent name, and version, along with individual JSON files for each tool describing its name, capability, description, and parameters. A separate mappings file links specific model IDs to their matching provider. There is also a folder with pseudo-code specifications for edit tool implementations, covering operations like string replacement and patch application, plus a sources document tracking where each schema definition came from. A small web interface is included that reads a public API file to let you browse the collected schemas. The intended use, per the README, is to hand one of these schemas to your own agent and ask it to build a working tool that matches that exact shape. The project is written in TypeScript and the README itself is quite sparse, focused mainly on explaining the folder layout rather than providing extensive documentation or examples.
A reference repository collecting the exact tool schemas different AI coding agents use, so developers can match a model's expected tool shape precisely.
Mainly TypeScript. The stack also includes TypeScript, JSON.
License terms are not described in the shown README, check the repository's license file for exact terms.
Setup difficulty is rated easy, with roughly 5min to a first successful run.
Mainly developer.
This repo across BitVibe Labs
Verify against the repo before relying on details.