yao-positioning-skill
Generate evidence-aware positioning reports for personal IPs, courses, products, services, brands, or companies by combining positioning theory…
Skill by Yao Team
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/yaojingang-yao-positioning-skill
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: design, marketing, research
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Python package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/yao-positioning-skill
What's inside
- Skill name
yao-positioning-skill- Description
- Generate evidence-aware positioning reports for personal IPs, courses, products, services, brands, or companies by combining positioning theory, course-marketing analysis, user intent, competitor research, market demand, and verifiable advantages. Use when users ask for 定位、差异化分析、优势分析、心智定位、品类定位、竞品定位研究、定位报告或再定位建议. Do not use for slogan-only copywriting, generic marketing ideas, standalone market research without a positioning recommendation, or brand knowledge-base construction.
- License
- MIT
- Metadata
- author: Yao Team
- Instructions
- 58 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Files (37)
- LICENSE
- README.md
- SKILL.md
- interface.yaml · agents/
- LICENSE-echarts.txt · assets/
- NOTICE-echarts.txt · assets/
- echarts.min.js · assets/
- report-template.html · assets/
- manifest.json
- ai-workflow.md · references/
- competitor-research-method.md · references/
- course-marketing-method.md · references/
- demand-and-differentiation-method.md · references/
- html-report-spec.md · references/
- intake-and-readiness-gate.md · references/
- metric-dictionary.md · references/
- plain-language-positioning-method.md · references/
- report-contract.md · references/
- report-review-method.md · references/
- source-credibility-policy.md · references/
- theory-system.md · references/
- index.html · reports/positioning-skill-system-overview-2026-07-16/
- requirements.txt
- render-report.py · scripts/
- report_model.py · scripts/
- review-report.py · scripts/
- sanitize-package.py · scripts/
- schema_validator.py · scripts/
- validate-intake.py · scripts/
- validate-report-data.py · scripts/
- permission_policy.json · security/
- permission_policy.md · security/
- evidence-ledger.md · templates/
- intake-brief.schema.json · templates/
- intake-questionnaire.md · templates/
- positioning-report.md · templates/
- report-data.schema.json · templates/
SKILL.md (first lines)
# 定位分析与报告 ## Owned Job - 为个人、课程、产品、服务或品牌生成定位报告。 - 研究五类竞争参照。 - 比较 2 至 3 套方案并推荐或降级。 ## Workflow 1. 按 `references/intake-and-readiness-gate.md` 接受问答、附件或 URL,整理符合 `templates/intake-brief.schema.json` 的 `intake-brief.json`,运行 `python3 scripts/validate-intake.py intake-brief.json`。 2. 仅当校验返回 `intake_ready` 才开始检索;否则停止并使用返回的缺失项与示例,每轮追问 1 至 3 个问题。开始前回显对象、目的、范围、已收资料、自述项和待核验项。 3. 选择 `local`、`standard` 或 `deep` 研究模式。需要当前竞品、价格或市场事实时必须联网核验。 4. 读取 `references/theory-system.md`。个人 IP 或课程任务再读 `references/course-marketing-method.md`。 5. 按 `references/ai-workflow.md` 建立主张清单、竞争集合、研究计划和证据台账。 6. 按 `references/source-credibility-policy.md` 区分事实、自述、推断、假设和建议;来源必须通过权威性、主张适配、直接性、时效性、独立性和市场适配检查。 7. 按 `references/competitor-research-method.md` 研究竞争与心智位置。禁止把“未发现公开证据”写成“不存在”。 8. 按 `references/demand-and-differentiation-method.md` 分析需求、市场空位和 D6 优势,执行否决规则。 9. 通过诊断与推荐门槛后生成 2 至 3 套定位方案并压力测试,只推荐一套;不通过时只输出初步观察和补证计划。 10. 按 `references/plain-language-positioning-method.md` 生成通俗的心智快照,再进入可视化报告。 11. 先生成符合 `templates/report-data.schema.json` 的 JSON,再运行 `python3 scripts/validate-report-data.py positioning-report-data.json`。 12. 校验通过后运行 `python3 scripts/render-report.py positioning-report-data.json --out positioning-report`。 13. 运行 `python3 scripts/review-report.py positioning-report`,修复语义、图表、移动端、打印和跨格式回归;不得通过删证据或升级置信度消除警告。 14. 按 `references/report-contract.md` 核对 HTML、JSON、Markdown 的结论、指标和引用一致性。模板或图表变更再按 `references/report-review-method.md` 执行三视口与 A4 PDF 栅格化审校。 ## Downgrade Rules - 研究就绪门槛未通过:只补信息,不检索竞品、不生成方案。 - 诊断门槛未通过:只输出初步观察和补证计划。 - 无法访问外部资料且用户未提供竞品:进入本地证据模式,输出补证清单。 - 没有用户需求证据:不得宣称市场空位。 - 没有可证明优势:输出能力建设建议,不制造差异化标签。 - 涉及第一、唯一、领先或效果承诺:要求独立证据和合规审查。 ## Output Contract 默认交付: - `positioning-report.html`:离线可视化决策报告。 - `positioning-report-data.json`:唯一结构化事实源。 - `positioning-report.md`:无脚本降级版与审计记录。 先用一页心智快照说清结论,再提供需求、竞争、优势、方案、行动、风险、证据和验证计划。个人 IP 或课程增加四层一致性,课程再增加行动路径、4P 与课程机制。缺失数据展示“证据不足”,不得伪造图表或总分。 交付前 `review-report.py` 必须通过。来源时效和证据警告可保留,但必须在报告和审校结果中可见;零警告交付使用 `--strict`。 ## Reference Map - 理论:`references/theory-system.md`、`references/plain-language-positioning-method.md`、`references/course-marketing-method.md` - 执行:`references/intake-and-readiness-gate.md`、`references/ai-workflow.md` - 研究:`references/source-credibility-policy.md`、`references/competitor-research-method.md` - 分析:`references/demand-and-differentiation-method.md` - 交付:`references/report-contract.md`、`references/report-review-method.md`、`references/html-report-spec.md`、`references/metric-dictionary.md` - 渲染资产:`assets/report-template.html`、`assets/echarts.min.js` - 系统说明:`reports/positioning-skill-system-overview-2026-07-16/index.html` - 分发净化:`scripts/sanitize-package.py`(仅在重建 ZIP 后执行)
From yaojingang/yao-open-skills, MIT.
More from yaojingang/yao-open-skills
- 0.824score
yao-naval-almanack-skill
yaojingang
Improve career choices, service/productization, reusable income/ownership, long-term partnerships, attention allocation, and desire/social-comparison…
SkillAFreshness: A★ 1.3KClaude Code +9 - 0.781score
yao-cold-ledger-skill
Yao Team
分析或创作“冷账叙事”中文关系故事:用精确数字、账单或工序、物件回环、克制对白和留白承载情绪。用于点名冷账叙事,或明确要求以冷峻账本感拆解参考文本、写片段、短篇或长篇;排除普通润色、通用公众号写作、事实核查、财务账务、真实人物私密爆料及可识别原句或情节复刻。
SkillBFreshness: B★ 1.3KClaude Code +9 - 0.727score
yao-expert-skill
Yao Team
Create expert-level learning reports and tutorials from any domain, industry, technology, role, market, product idea, or vague field question.
SkillCFreshness: C★ 1.3KClaude Code +9 - 0.727score
yao-gametheory-skill
yaojingang
Route CEO-facing competition, negotiation, channel, pricing, platform, M&A, financing, competitor-response, alliance, and regulator problems through an…
SkillCFreshness: C★ 1.3KClaude Code +9 - 0.727score
yao-interpreter-skill
Yao Team
Analyze local Agent Skill folders, SKILL.md files, or safe zip archives as untrusted input; explain purpose, trigger contract, structure, usage readiness…
SkillCFreshness: C★ 1.3KClaude Code +9 - 0.727score
yao-kelly-skill
yaojingang
Turn uncertain resource-allocation requests into practical action plans using Kelly sizing as a conservative allocation engine.
SkillCFreshness: C★ 1.3KClaude Code +9 - 0.727score
yao-open-skills-sync
yaojingang
Manage the yao-open-skills public collection.
SkillCFreshness: C★ 1.3KClaude Code +9 - 0.727score
yao-tutorial-skill
yaojingang
Create standalone beginner tutorial packages from a topic or supplied references, with adaptive research, course-style outline design, chapter visuals, and…
SkillCFreshness: C★ 1.3KClaude Code +9 - 0.727score
yao-websecurity-skill
yaojingang
Use when auditing an authorized website, SaaS, API, AI app, local code path, GitHub repo, staging URL, or owned runtime for security risks, vulnerability…
SkillCFreshness: C★ 1.3KClaude Code +9 - 0.727score
yao-copyright-skill
yaojingang
Use when adding or auditing Yao copyright comment headers for an agent skill package, especially SKILL.md, README.md, references, prompts, templates…
SkillCFreshness: C★ 1.3KClaude Code +9 - 0.727score
yao-crux-skill
Yao Team
Use when a user is stuck in a complex real-world situation with many conflicts, limited resources, unclear priorities, or asks to identify the principal…
SkillCFreshness: C★ 1.3KClaude Code +9 - 0.727score
yao-demand-skill
Yao Team
Evaluate product demand from product links, product descriptions, PRDs, websites, app-store pages, white papers, sales decks, screenshots, or funding…
SkillCFreshness: C★ 1.3KClaude Code +9
Signals
- Stars
- 1,324
- Forks
- 144
- Open issues
- 0
- Last commit (this folder)
- Jul 16, 2026 (3 months ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/yaojingang/yao-open-skills
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from Yao Team
- 0.781score
yao-cold-ledger-skill
Yao Team
分析或创作“冷账叙事”中文关系故事:用精确数字、账单或工序、物件回环、克制对白和留白承载情绪。用于点名冷账叙事,或明确要求以冷峻账本感拆解参考文本、写片段、短篇或长篇;排除普通润色、通用公众号写作、事实核查、财务账务、真实人物私密爆料及可识别原句或情节复刻。
SkillBFreshness: B★ 1.3KClaude Code +9 - 0.727score
yao-crux-skill
Yao Team
Use when a user is stuck in a complex real-world situation with many conflicts, limited resources, unclear priorities, or asks to identify the principal…
SkillCFreshness: C★ 1.3KClaude Code +9 - 0.727score
yao-demand-skill
Yao Team
Evaluate product demand from product links, product descriptions, PRDs, websites, app-store pages, white papers, sales decks, screenshots, or funding…
SkillCFreshness: C★ 1.3KClaude Code +9 - 0.727score
yao-interpreter-skill
Yao Team
Analyze local Agent Skill folders, SKILL.md files, or safe zip archives as untrusted input; explain purpose, trigger contract, structure, usage readiness…
SkillCFreshness: C★ 1.3KClaude Code +9 - 0.727score
yao-expert-skill
Yao Team
Create expert-level learning reports and tutorials from any domain, industry, technology, role, market, product idea, or vague field question.
SkillCFreshness: C★ 1.3KClaude Code +9
Cite this page
yao-positioning-skill. Agent Mart. https://agentmart.dev/skills/yaojingang-yao-positioning-skill (last updated 2026-10-24)