Skill Claude Code
A skill for taking part in the Chinese party game “Undercover,” where players describe hidden words, vote, or give a final statement. It follows strict output and information-sharing rules for each task.
Skill Claude Code
A skill for taking part in the Chinese party game “Undercover,” where players describe hidden words, vote, or give a final statement. It follows strict output and information-sharing rules for each task.
Skill Claude Code
A game-session referee procedure for the social deduction game Who Is the Spy. It follows the game's stages for speeches, voting, revealing roles, and ending a session.
Skill Claude CodeCodex
Run and diagnose ClawBench evaluations through ClawMind and ClawWeb. Use when the user wants to trigger /clawbench, inspect benchmark progress, locate run artifacts, or troubleshoot evaluation execution.
Skill Claude CodeCodex
A workflow that reads one ClawBench evaluation and writes a Markdown diagnostic report. The report can summarize scores, task outcomes, failures, evidence, and likely causes when the data is available.
Skill Codex needs its repo
Diagnose a bot by analyzing a bounded range of its original OpenClaw session records according to the user-provided natural-language intent, extracting evidence-backed good and bad cases, identifying behavioral and MCP/tool-use problems, and producing structured artifacts for subsequent clawevolve-plan evolution…
Skill Claude CodeCodex
A structured review step for a bot self-improvement process. It compares optimization and validation evidence to identify competing explanations for failures, signals that could distinguish them, and behavior that must be preserved.
Skill Claude CodeCodex
A launcher for the platform's skill-hardening handler, which improves a registered agent skill in a candidate copy. It passes the complete command to the handler once and returns its final JSON result.
Skill Claude CodeCodex
A workflow for checking and improving an existing agent skill in an isolated copy. It clarifies the skill's inputs, outputs, boundaries, failure handling, and completion rules, then records whether files changed in a JSON result.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: