Career Planning Claude Skill — Development skill for Claude Code
A Claude Code skill for structured, anti-hallucination career & employment planning (职业规划/就业规划/转行).
How to install Career Planning Claude Skill
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open SparkySunny/career-planning-claude-skill and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Career Planning Claude Skill does
A Claude Code skill for structured, anti-hallucination career & employment planning (职业规划/就业规划/转行). RIASEC, Schein Career Anchors, SWOT, GROW, SMART. Frameworks fixed in reference files; no fabricated salary/market data.
Alternatives in Development
- Career-Ops For Codex — Read CLAUDE.md for all project instructions, routing, and behavioral rules 32.5k ★
- Swot Analysis — SWOT analysis with actionable recommendations per quadrant 7.8k ★
- Memory Gc — Inspect or rotate the autopilot ledger JSONL files (findings.jsonl, negatives.jsonl) 3.8k ★
README
Career Planning · 职业规划 / 就业规划 — Claude Skill
A [Claude Code](https://docs.claude.com/en/docs/claude-code) skill for structured, **anti-hallucination** career & employment planning (职业规划 / 就业规划 / 转行 / 选专业 / 求职方向).
中文 · [English](#english)
中文
这是一个给 Claude Code 用的 **职业规划 / 就业规划** 技能(skill)。它不给一句"你应该去做产品经理"的判决,而是带用户走一遍结构化流程,产出 **3–5 个候选方向**、**加权决策矩阵**、和**带时间节点的行动计划**,让用户自己拍板。
**最大特点:反幻觉(anti-hallucination)。** 职业建议最容易捏造两类东西——框架细节(凭记忆说错 RIASEC / 职业锚)和市场数字(凭空编一个薪资区间)。本技能把所有框架固定在 reference 文件里照着读、把市场数据规则写死(查不到来源就不给数字、改指向可查来源),从结构上防止 AI 误导重大决策。
框架(均有出处,详见 references/frameworks.md)
- RIASEC / Holland Codes(霍兰德) — 兴趣六类型 + 3 字母码
- Schein Career Anchors(施恩职业锚) — 8 类职业锚
- SWOT — 严格区分内部(优劣势)/ 外部(机会威胁)
- GROW(Whitmore)+ SMART — 目标与行动
- 加权决策矩阵 + Ikigai(诚实标注:四圈图是西方化流行版本,非传统日本概念)
工作流
- Intake 信息收集 → 2. 自我评估(RIASEC + 职业锚 + SWOT)→ 3. 生成 3–5 候选方向(每个含:简述 / 为何契合 / 主要权衡 / 现实切入路径)→ 4. 加权决策矩阵 → 5. GROW + SMART 行动计划(1/3/6/12 月 + 技能缺口分析)→ 6. 复盘。流程可按问题深度缩放——快问题走轻量版,但反捏造铁律不可跳过。
文件结构
career-planning/
├── SKILL.md # 主技能文件(工作流 + 输出模板 + 反捏造三铁律)
├── references/
│ ├── frameworks.md # 框架定义与出处(反幻觉锚 · 理论)
│ ├── data-sources.md # 劳动市场数据来源与引用规则(反幻觉锚 · 事实)
│ ├── intake.md # 结构化信息收集问卷
│ └── career-mapping.md# RIASEC + 锚 → 候选方向的映射方法
├── evals/
│ └── evals.json # 3 个测试用例与期望输出
├── README.md
└── LICENSE
安装 / 使用
把 `career-planning/` 整个目录放进 Claude Code 的 skills 目录(`~/.claude/skills/`,Windows 下为 `C:\Users\<你>\.claude\skills\`)。然后对 Claude 说"帮我做职业规划 / 就业规划 / 转行分析 / 选方向"即可触发。用中文提问就用中文回答。
测试结果(with-skill vs 无技能 baseline)
3 个用例全部通过 expected_output 判据。关键对比——eval-3"花两万报三个月 Python 培训班"用例:
- 无技能 baseline:凭空写出"薪资 6K-10K""一个岗位收到几百份简历"(无来源、无年份、无地区)。
- 本技能 with-skill:明确拒编,该出数字处均写"暂无核实数据",并指向 国家统计局 / 智联招聘《中国企业招聘薪酬报告》/ BOSS直聘 / O*NET,提示查询须带年份与地区。
即:**证实防住了最易捏造的"市场数据幻觉"**——正是本技能的设计目标。
English
A **career-planning / employment-planning** skill for Claude Code. It walks a user
Related Skills
Hallucination Check
Catch invented APIs, fabricated signatures, and non-existent imports — cross-referenced against your actually-
Grounded Research Skill
Anti-hallucination research skill for Claude Code — admits uncertainty, extracts direct quotes before analysis
Paper Trail New Paper
Start writing an academic paper (IMRaD structure) on a topic, with anti-hallucination citation verification at
Agent Project Planner
Dispatch hub:project-planner — Smart project planning agent.
MCP Edit
tilth_write replaces the host Edit and Write tools. Its full usage — hash/overwrite/append modes, hash anchors
Hashline
Stable line-addressed file editing for Claude Code, Codex, ... using content-hashed anchors instead of fragile
Related Agents
CzechLawAdvisor
Czech labor law advisor -- zákoník práce, anti-discrimination, mandatory posting elements, GDPR compliance. US
Code Sentinel
Use this agent as the engineering discipline enforcer and production-quality sentinel — dispatched to audit AN
Legal Counsel
Analyze legal issues, review contracts, and draft clear legal work product with rigorous anti-hallucination an