majiayu000
majiayu000/claude-skill-registry
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/majiayu000/claude-skill-registry --skill agents命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
LLM token logprobs and calibration. Per-decision confidence, ECE, Brier, reliability diagrams, low-confidence triage.
Analyze LLM token logprobs and calibration. Use for per-decision confidence, ECE, Brier scores, reliability diagrams, and low-confidence triage.
回顾最近 N 天的 Claude Code 使用记录——扫描原始会话数据，按主题分组汇总"我都做了什么"，并从个人操作系统视角输出模式、风险与增删建议。当用户说 /recap、"看看我这几天做了什么"、"回顾一下我最近的会话"、"这两天我用 claude 干了啥"、"活动回顾" 时使用。
基于 SOC 职业分类
正在显示 SKILL.md
| name | agents |
| description | List all available agents (core + expert) |
| user-invocable | true |
List all agents available in Director Mode Lite.
Purpose: Code quality, security, best practices review. Triggers: "review", "check code", before commits
Purpose: Systematic debugging for errors. Triggers: Errors, test failures, "bug", "debug"
Purpose: Documentation creation and maintenance. Triggers: New features, "document", "README"
CLAUDE.md design patterns and best practices.
MCP configuration and troubleshooting.
Custom agent creation and configuration.
Custom skill/command creation.
Automation hooks and triggers.
"Use code-reviewer to check src/auth/"
"I need the debugger - tests are failing"
"Have doc-writer update the API docs"
---
name: my-agent
description: What this agent does
tools: Read, Grep, Glob, Bash
---
# Agent Name
## When to Activate
## Process
## Output Format
Save to .claude/agents/my-agent.md