awesome-reverse-engineered
Search 17,000+ open-source projects built by reverse engineering closed apps, APIs, devices, protocols, file formats and games before you start. Ranked by stars, language, license and activity; reads the matching README. Standard library only.
Documentation
reverse-engineered MCP server
MCP server for awesome-reverse-engineered: 17,000+ open-source projects that exist because someone reverse engineered a closed app, API, device, protocol, file format, game or AI product.
Don't reverse engineer it twice: your agent searches the index before it starts.
Tools
| Tool | What it does |
|---|---|
search_reverse_engineered | Ranked search by product, vendor, protocol or task (tuya local, psd parser, n64 decomp). Filters: category, language, minimum stars, archived. |
get_reverse_engineered_project | One entry (stars, license, last push, archived) plus the start of its README. |
list_reverse_engineered_categories | Categories and subcategories with counts. |
Install
Needs uv. No other dependencies (Python standard library only).
Claude Code:
claude mcp add reverse-engineered -- uvx --from "git+https://github.com/alexgaoth/awesome-reverse-engineered#subdirectory=mcp" reverse-engineered-mcp
Claude Desktop, Cursor and other clients:
{
"mcpServers": {
"reverse-engineered": {
"command": "uvx",
"args": ["--from", "git+https://github.com/alexgaoth/awesome-reverse-engineered#subdirectory=mcp", "reverse-engineered-mcp"]
}
}
}
With Docker (image ghcr.io/alexgaoth/reverse-engineered-mcp):
claude mcp add reverse-engineered -- docker run -i --rm ghcr.io/alexgaoth/reverse-engineered-mcp
Without uv: download reverse_engineered_mcp.py and re_index.py into one folder and run python3 reverse_engineered_mcp.py.
Data
The server downloads data/entries.json (CC0) on first use and caches it for 24 hours in ~/.cache/awesome-reverse-engineered/. Set RE_INDEX_DATA=/path/to/entries.json to use a local copy. get_reverse_engineered_project reads READMEs through the GitHub API; set GITHUB_TOKEN for a higher rate limit.
re_index.py is a copy of the agent skill's script; scripts/render.py keeps it in step.
Releases
.github/workflows/mcp-publish.yml builds the image, pushes it to GHCR and publishes server.json to the MCP Registry as io.github.alexgaoth/awesome-reverse-engineered. The version comes from server.json.