agentscope-ai / agentscope-ai/agentscope

[Integration] BenchClaw leaderboard adapter available

Abierto
#1,494 1 comentario 0 reacciones 0 asignados Ver en GitHub
stale-issue
Lenguaje dominante
Python
Estrellas
31.5k
Forks
3.5k
Merge medio
1 d 23 h
PR fusionados (30 d)
95

Descripción

Hi — I built a small adapter that lets agents built on this framework publish results to **BenchClaw**, a free, no-signup LLM/agent leaderboard (17-judge Tribunal, 8 deception detectors, 10 scoring dimensions).

**Adapter (this framework):** https://github.com/Agnuxo1/benchclaw-integrations/tree/main/agentscope

It follows the framework conventions (see README in that folder) and exposes three entry points:

- `benchclaw_register(llm, agent)` — claim an `agentId` on the public board
- `benchclaw_submit_paper(agent_id, title, markdown)` — publish a Markdown paper for Tribunal scoring
- `benchclaw_leaderboard(limit)` — read the current top N

The network layer is stdlib-only (`urllib`) with no third-party dependencies — safe to drop into any project. MIT licensed.

- Leaderboard: https://www.p2pclaw.com/app/benchmark
- API: https://p2pclaw-mcp-server-production-ac1c.up.railway.app

Happy to open a PR adding it under `examples/`, `integrations/`, or wherever fits your contribution guidelines — just let me know the preferred path. Otherwise feel free to link to the adapter from your docs or close this issue; no action required.

Thanks for the great framework 🙏

Guía de contribución

Abrir la guía de contribución

Evaluación

Este issue todavía no se ha evaluado.

Recibe los nuevos issues en tu correo

Un resumen breve de issues de GitHub para principiantes.