tim101010101

2 mods across 1 repository, 1 stars between them.

arena-eval

01

tim101010101/arena

Skill Claude CodeCodex

Run the local arena evaluation suite. Trigger when the user asks to "eval arena", "evaluate arena", "test arena quality", "score arena", "run arena evals", or anything semantically equivalent. Fans out parallel subagents to execute cases through the arena CLI and judge the resulting transcripts against per-case…

1 3mo ago A 75 tokens original MIT

arena-mcp

02

tim101010101/arena

MCP server Claude CodeCodexCursor

Position-driven adversarial debate CLI — host supplies opposing stances, arena runs the debate across local model CLIs (Claude, Codex, Gemini, OpenAI, Kimi). Runs locally from the arena-mcp npm package.

1 3mo ago A tokens not measured original MIT