基于 SOC 职业分类
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 task-classifier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。
正在显示 SKILL.md
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。
| name | task-classifier |
| description | Analyzes task descriptions and classifies them into categories for agent selection |
| version | 1.0.0 |
| tags | ["classification","agent-selection","automation"] |
Analyzes task descriptions using keyword matching to suggest appropriate agent specializations.
./.claude/skills/task-classifier/classify-task.sh "Task description"
| Category | Keywords | Use Case |
|---|---|---|
| frontend | ui, ux, react, component, css, styling, layout, responsive, interface | UI/UX development |
| backend | api, endpoint, server, database, rest, graphql, service, authentication | Server-side development |
| devops | docker, kubernetes, ci/cd, deployment, infrastructure, container, pipeline | Infrastructure work |
| testing | test, qa, validation, coverage, integration, unit, e2e | Quality assurance |
| security | security, auth, encryption, vulnerability, audit, penetration | Security work |
| data | database, sql, migration, schema, data, model, entity | Data modeling |
| performance | performance, optimization, speed, cache, memory, cpu | Performance tuning |
| general | (default) | General development |
Comma-separated list of classifications:
$ classify-task.sh "Create a React dashboard with API integration"
frontend,backend,testing
This skill is typically used with cfn-agent-selector to determine which agents to spawn:
CLASSIFICATION=$(classify-task.sh "$TASK_DESCRIPTION")
AGENTS=$(select-agents.sh --classification "$CLASSIFICATION" --mode standard)
# Frontend task
$ classify-task.sh "Build responsive navigation component"
frontend
# Full-stack task
$ classify-task.sh "Create REST API with React admin panel"
frontend,backend
# DevOps task
$ classify-task.sh "Setup CI/CD pipeline with Docker"
devops
# Security audit
$ classify-task.sh "Perform security audit and fix vulnerabilities"
security
grep -E for case-insensitive pattern matchingcfn-v3-coordinator - For automatic agent selectioncfn-agent-selector - As input for agent mapping