# Agent Skills Phoebe Picks Wishlist

This is the working queue for Skills Phoebe wants to explore, test with a real artifact, and either publish to the gallery or reject quietly.

The public gallery should only show Skills that feel useful enough to recommend. Rejected Skills can stay here as learning notes without becoming public recommendations.

## Status Key

| Status | Meaning |
| --- | --- |
| Wishlist | Worth trying, not tested yet. |
| Exploring | Currently being used in a real build or evaluation. |
| Published | Useful enough to add to the public gallery. |
| Rejected | Tried, but not worth recommending right now. |

## Evaluation Queue

| Skill | Source / reference | Stars snapshot | Platform fit | Scenario | Ground-up build target | 80% concept coverage | Rating /10 | Date explored | Status | Gallery link |
| --- | --- | --- | --- | --- | --- | --- | --- | --- | --- | --- |
| `frontend-design` | `anthropics/skills`, `skills/frontend-design` | 161,869 on 2026-07-17 | Claude Code, Codex, Hermes AI, Cursor, Gemini CLI | Product design, frontend, UI | Signal Desk product screen | 6/6 key concepts visible | 8.6/10 | 2026-07-17 | Published | `demos/product-design-studio/index.html#frontend-design` |
| `canvas-design` | `anthropics/skills`, `skills/canvas-design` | 161,869 on 2026-07-17 | Claude Code, Codex, Hermes AI, Cursor, Gemini CLI | Product design, visual philosophy | Quiet Leverage philosophy and SVG canvas | 6/7 key concepts visible; SVG/HTML used instead of binary PDF/PNG for GitHub Pages | 7.8/10 | 2026-07-17 | Published | `demos/product-design-studio/index.html#canvas-design` |
| `emil-design-eng` | `emilkowalski/skills`, `skills/emil-design-eng` | 15,002 on 2026-07-17 | Claude Code, Codex, Hermes AI, Cursor, Gemini CLI | Product design, micro-interactions | Product interaction polish lab | 8/9 key concepts visible; drag gesture physics skipped | 8.4/10 | 2026-07-17 | Published | `demos/product-design-studio/index.html#emil-design-eng` |
| `design-taste-frontend` | `Leonxlnx/taste-skill`, `skills/taste-skill/SKILL.md` | 78,263 on 2026-08-20 | Claude Code, Codex, Hermes AI, Cursor, Gemini CLI | Visual taste, anti-slop review, redesign safety | Design Taste Frontend Lab v2 field test | 13/15 concept families exercised; advanced GSAP and framework-specific packages skipped because the proof is intentionally static | 8.8/10 | 2026-08-20 | Published | `demos/design-taste-frontend-studio/index.html` |
| `design-review` | `garrytan/gstack`, Open Design wrapper `skills/design-review` | 122,177 on 2026-07-16 | Claude Code, Codex, Hermes AI | Design critique, QA | Critique Loop before-after audit board | 5/5 key concepts visible | 8.0/10 | 2026-07-16 | Published | `demos/design-review-critique-loop/index.html` |
| `high-end-visual-design` | `Leonxlnx/taste-skill`, Open Design wrapper `skills/soft-skill` | 64,018 on 2026-07-16 | Claude Code, Codex, Hermes AI | Typography, layout polish, visual craft | High-End Visual Design Atelier | 8/10 key concepts visible; full scroll choreography and full-screen menu sequencing kept lighter for this static gallery | 7.9/10 | 2026-07-16 | Published | `demos/high-end-visual-design-showcase/index.html` |
| `design-consultation` | `garrytan/gstack`, `gstack design-consultation skill` | 122,177 on 2026-07-16 | Claude Code, Codex, Hermes AI | Brand system, tokens, design principles | System Room design-system workshop | 5/5 key concepts visible | 8.1/10 | 2026-07-16 | Published | `demos/design-consultation-system-room/index.html` |
| `impeccable` | `pbakaus/impeccable`, `.agents/skills/impeccable/SKILL.md` | 47,939 on 2026-07-19 | Codex, Claude Code, Cursor, Gemini CLI, opencode | Design direction, frontend craft, critique, audit | Coastal Circuit × Soft Radical homepage redesign | 11/12 concept clusters exercised; live asset slicing skipped because semantic HTML/CSS and existing artifact screenshots covered the approved mock | 9.0/10 | 2026-07-19 | Published | `index.html#top` |
| `pm-skills` | `phuryn/pm-skills`, 65 Skills across 8 product domains | 24,430 on 2026-07-25 | Claude Code, Codex, Hermes AI, Cursor, Gemini CLI, opencode | Product management, research, discovery, strategy, execution, analytics, growth, GTM, toolkit | Lantern Product OS | 65/65 installed Skills exercised, 100% package coverage | 8.9/10 | 2026-07-25 | Published | `demos/pm-decision-room/index.html` |
| `find-skills` | `vercel-labs/skills`, `skills/find-skills/SKILL.md`; #1 all-time on Skills.sh with 2.7M installs | 27,182 on 2026-07-25 | Codex, Claude Code, Hermes AI, Cursor, Gemini CLI, GitHub Copilot, Windsurf | Agent workflow, research, Skill discovery | Skill Discovery Desk + 20-Skill research pass | 6/6 concepts exercised; direct CLI execution remains optional pending explicit trust | 7.8/10 | 2026-07-25 | Published | `demos/find-skills-discovery-desk/index.html` |
| `d3-visualization` | `jiannanya/snow-d3`, `SKILL.md`; catalog wrapper in `nexu-io/open-design`, `skills/d3-visualization/SKILL.md` | Canonical source 4 stars; wrapper repo 81,586 on 2026-07-26 | Codex, Claude Code, Hermes AI, Cursor, Gemini CLI | Data visualization, data, reporting, frontend | Skill Constellation | 9/10 concept families; geography skipped because the dataset has no spatial dimension | 8.3/10 | 2026-07-26 | Exploring; canonical source below credibility gate | `demos/d3-skill-constellation/index.html` |
| `dbt-transformation-patterns` | `wshobson/agents`, `plugins/data-engineering/skills/dbt-transformation-patterns/SKILL.md` | 38,467 on 2026-08-04 | Codex, Claude Code, Cursor, Gemini CLI, GitHub Copilot, Windsurf | Data engineering, analytics engineering | Reliable Data Pipeline Lab dbt project and build simulator | 9/10 concept families; live warehouse execution replaced by inspectable project files and browser simulation | 8.6/10 | 2026-08-04 | Published | `demos/reliable-data-pipeline-lab/index.html#dbt-lab` |
| `airflow-dag-patterns` | `wshobson/agents`, `plugins/data-engineering/skills/airflow-dag-patterns/SKILL.md` | 38,467 on 2026-08-04 | Codex, Claude Code, Cursor, Gemini CLI, GitHub Copilot, Windsurf | Data engineering, orchestration | Reliable Data Pipeline Lab DAG and recovery simulator | 9/10 concept families; dynamic multi-DAG generation skipped for one bounded data product | 8.4/10 | 2026-08-04 | Published | `demos/reliable-data-pipeline-lab/index.html#airflow-lab` |
| `customer-research` | `coreyhaines31/marketingskills`, `skills/customer-research/SKILL.md` | 43,591 on 2026-08-13 | Codex, Claude Code, Cursor, Gemini CLI, opencode | Customer research, KOL evidence | KOL Evidence Studio Signal Desk and research synthesis | 8/9 concept families; direct interviews and evidence-based personas skipped because no first-party participants were supplied | 9.0/10 | 2026-08-13 | Published | `demos/kol-evidence-studio/index.html#customer-research` |
| `product-marketing` | `coreyhaines31/marketingskills`, `skills/product-marketing/SKILL.md` | 43,591 on 2026-08-13 | Codex, Claude Code, Cursor, Gemini CLI, opencode | Positioning, KOL context | KOL Evidence Studio Positioning Room and canonical marketing context | 11/12 context sections; pricing skipped for a free knowledge product and human confirmation remains pending | 8.7/10 | 2026-08-13 | Published | `demos/kol-evidence-studio/index.html#product-marketing` |
| `content-strategy` | `coreyhaines31/marketingskills`, `skills/content-strategy/SKILL.md` | 43,591 on 2026-08-13 | Codex, Claude Code, Cursor, Gemini CLI, opencode | Content strategy, KOL editorial system | KOL Content Engine Strategy Map | 9/10 concept families; live keyword-volume and competitor-gap datasets were unavailable | 8.8/10 | 2026-08-13 | Published | `demos/kol-content-engine/index.html#content-strategy` |
| `social` | `coreyhaines31/marketingskills`, `skills/social/SKILL.md` | 43,591 on 2026-08-13 | Codex, Claude Code, Cursor, Gemini CLI, opencode | Social strategy, KOL distribution | KOL Content Engine Distribution Desk and Learning Loop | 10/11 concept families; live listening and a 500-post creator dataset require platform data | 8.6/10 | 2026-08-13 | Published | `demos/kol-content-engine/index.html#social` |
| `copywriting` | `coreyhaines31/marketingskills`, `skills/copywriting/SKILL.md` | 43,591 on 2026-08-13 | Codex, Claude Code, Cursor, Gemini CLI, opencode | Conversion copy, landing pages | Proof to Action Lab conversion page and copy annotations | 9/10 concept families; a separate copy-editing pass remains a future field test | 8.8/10 | 2026-08-13 | Published | `demos/proof-to-action-lab/index.html#copywriting` |
| `analytics` | `coreyhaines31/marketingskills`, `skills/analytics/SKILL.md` | 43,591 on 2026-08-13 | Codex, Claude Code, Cursor, Gemini CLI, opencode | Measurement, event contracts | Proof to Action Lab measurement inspector and tracking plan | 10/11 concept families; live GA4/GTM publication requires owned accounts and consent | 8.7/10 | 2026-08-13 | Published | `demos/proof-to-action-lab/index.html#analytics` |
| `lead-magnets` | `coreyhaines31/marketingskills`, `skills/lead-magnets/SKILL.md` | 44,208 on 2026-08-14 | Codex, Claude Code, Cursor, Gemini CLI, opencode | Lead magnets, educational resources | Agent Skill Evaluation Starter Kit checklist and offer strategy | 10/11 concept families; paid amplification excluded without a budget or first-party baseline | 8.8/10 | 2026-08-14 | Published | `demos/skill-evaluation-starter-kit/index.html#lead-magnets` |
| `emails` | `coreyhaines31/marketingskills`, `skills/emails/SKILL.md` | 44,208 on 2026-08-14 | Codex, Claude Code, Cursor, Gemini CLI, opencode | Email nurture, lifecycle automation | Agent Skill Evaluation Starter Kit five-email learning journey | 10/11 concept families; live sending and deliverability testing require a consented list and platform | 8.7/10 | 2026-08-14 | Published | `demos/skill-evaluation-starter-kit/index.html#emails` |
| `grill-me` | `mio-openliven/codex-grill-me-skill` has 2 stars; likely alternate `RobMitt/grill-me-skill` has 123 stars | Below 10k source gate on 2026-07-16 | Codex first; Claude Code and Hermes AI if adapted | Requirement grilling, decision pressure-test | Field test completed, public demo removed | Method was useful, but source credibility gate failed | N/A - not published | 2026-07-14 | Rejected | Removed from public gallery |
| `using-superpowers` | `obra/superpowers`, `skills/using-superpowers` | 255,624 on 2026-07-16 | Claude Code, Codex, Cursor, Gemini CLI, opencode, Kimi; Hermes AI if adapted | Agent workflow, skill discipline | Build a new feature through the Superpowers process from zero | Pending | Pending | 2026-07-14 | Exploring | Pending |
| `brainstorming` | `obra/superpowers`, `skills/brainstorming` | 255,624 on 2026-07-16 | Claude Code, Codex, Cursor, Gemini CLI, opencode, Kimi; Hermes AI if adapted | Requirements, design thinking, spec writing | Produce a ground-up spec and artifact plan for one new gallery feature | Pending | Pending | 2026-07-14 | Exploring | Pending |

## Source-Backed Discovery Queue

This queue contains 105 verified Skill files; 92 remain on the wishlist after `dashboard` was promoted on 2026-07-28, `algorithmic-art` plus `data-report` on 2026-08-02, `dbt-transformation-patterns` plus `airflow-dag-patterns` on 2026-08-04, `customer-research`, `product-marketing`, `content-strategy`, `social`, `copywriting`, plus `analytics` on 2026-08-13, and `lead-magnets` plus `emails` on 2026-08-14. The original 20 were checked on 2026-07-25; the knowledge-base, Office, data-to-dashboard, marketing, one-person-company, design, data, ML, LLM, agent, product, content-writing, WeRead, and Chinese-medicine research collections were checked on 2026-07-26. A new batch of 20 emerging data, visualization, data-science, and ML Skills was verified on 2026-08-09. Repository popularity and Skills.sh adoption are discovery signals, not recommendations: every remaining candidate stays `Wishlist` until its source is read fully and a ground-up build exercises about 80% of the method.

The 2026-08-09 batch was ranked for a balanced combination of source credibility, active maintenance, technical relevance, distinct field-test value, and momentum. Nineteen candidates come from the actively maintained 33,024-star `K-Dense-AI/scientific-agent-skills` repository. `chart-visualization` is an emerging exception: its AntV source had 455 stars but about 5.2K Skills.sh installs, so it remains a wishlist experiment and does not yet pass the normal repository-star publication gate.

Chinese-medicine entries are research and knowledge-management candidates only. They must not provide diagnosis, treatment, dosage, or clinical decision support, and their low-star sources require additional credibility, domain-expert, licensing, and safety review before any public recommendation.

| Rank | Skill | Verified source path | Repo stars | Focus | Why it is on the wishlist | Status |
| --- | --- | --- | ---: | --- | --- | --- |
| 1 | `algorithmic-art` | `anthropics/skills`, `skills/algorithmic-art/SKILL.md` | 165,758 | Design, generative art | Built Evidence Tides with a five-paragraph philosophy, deterministic particles, meaningful controls, seed navigation, reset, regeneration, and PNG export. | Published |
| 2 | `xlsx` | `anthropics/skills`, `skills/xlsx/SKILL.md` | 164,075 | Data analysis, spreadsheets | Test formula-safe spreadsheet analysis, formatting, charts, and recalculation QA. | Wishlist |
| 3 | `skill-creator` | `anthropics/skills`, `skills/skill-creator/SKILL.md` | 164,075 | AI agents, Skill engineering | Build and evaluate a new Skill with prompt tests, quantitative checks, and iteration. | Wishlist |
| 4 | `mcp-builder` | `anthropics/skills`, `skills/mcp-builder/SKILL.md` | 164,075 | LLM tools, AI agents | Build a narrowly scoped MCP server and judge tool design by task completion quality. | Wishlist |
| 5 | `artifacts-builder` | `nexu-io/open-design`, `skills/artifacts-builder/SKILL.md` | 81,448 | Product design, interactive artifacts | Test a multi-component product artifact with a real information architecture and interaction model. | Wishlist |
| 6 | `dashboard` | `nexu-io/open-design`, `design-templates/dashboard/SKILL.md` | 82,010 | Dashboard design, information hierarchy | Built the Skill Portfolio Command Center with 10/10 concept families, inline SVG, KPI hierarchy, search states, and responsive QA. | Published |
| 7 | `data-report` | `nexu-io/open-design`, `skills/data-report/SKILL.md` | 83,147 | Data analytics, reporting | Built Skill Portfolio Evidence Report from the real review catalog with KPIs, fixed-height charts, source rows, computed insights, and methodology. | Published |
| 8 | `data-storytelling` | `wshobson/agents`, `plugins/business-analytics/skills/data-storytelling/SKILL.md` | 38,211 | Data analytics, executive communication | Test whether a setup–conflict–resolution structure improves an executive decision. | Wishlist |
| 9 | `kpi-dashboard-design` | `wshobson/agents`, `plugins/business-analytics/skills/kpi-dashboard-design/SKILL.md` | 38,211 | BI, dashboards, product analytics | Build a governed KPI dashboard with consistent definitions, hierarchy, and action thresholds. | Wishlist |
| 10 | `llm-evaluation` | `wshobson/agents`, `plugins/llm-application-dev/skills/llm-evaluation/SKILL.md` | 38,211 | LLM evaluation, ML | Compare automated metrics, human review, regression tests, and A/B evidence on one LLM task. | Wishlist |
| 11 | `rag-implementation` | `wshobson/agents`, `plugins/llm-application-dev/skills/rag-implementation/SKILL.md` | 38,211 | RAG, LLM applications | Build a source-grounded retrieval prototype and measure retrieval and answer quality separately. | Wishlist |
| 12 | `agentic-eval` | `github/awesome-copilot`, `skills/agentic-eval/SKILL.md` | 37,023 | AI agents, evaluation | Test generate–evaluate–critique–refine loops against a fixed rubric and stopping rule. | Wishlist |
| 13 | `autoresearch` | `github/awesome-copilot`, `skills/autoresearch/SKILL.md` | 37,023 | AI agents, ML experimentation | Run a bounded autonomous experiment loop with a baseline, measurable target, and keep/discard log. | Wishlist |
| 14 | `exploratory-data-analysis` | `K-Dense-AI/scientific-agent-skills`, `skills/exploratory-data-analysis/SKILL.md` | 31,724 | Data science, EDA | Audit missingness, leakage, outliers, and transformation sensitivity on an authorized dataset. | Wishlist |
| 15 | `scientific-visualization` | `K-Dense-AI/scientific-agent-skills`, `skills/scientific-visualization/SKILL.md` | 31,724 | Data science, visual communication | Produce an accessible publication-ready figure without hiding uncertainty or missing data. | Wishlist |
| 16 | `statsmodels` | `K-Dense-AI/scientific-agent-skills`, `skills/statsmodels/SKILL.md` | 31,724 | Statistics, ML, econometrics | Test model selection, diagnostics, residual checks, uncertainty, and reproducible inference. | Wishlist |
| 17 | `Product Manager` | `alirezarezvani/claude-skills`, `.gemini/skills/product-manager/SKILL.md` | 23,170 | Product management | Turn a vague stakeholder request into a short outcome-led spec, explicit trade-off, and success metric. | Wishlist |
| 18 | `product-discovery` | `alirezarezvani/claude-skills`, `product-team/skills/product-discovery/SKILL.md` | 23,170 | Product discovery | Test opportunity mapping, assumption ranking, interviews, prototypes, and evidence-based decision gates. | Wishlist |
| 19 | `product-analytics` | `alirezarezvani/claude-skills`, `product-team/skills/product-analytics/SKILL.md` | 23,170 | Product analytics, data | Build a metric hierarchy with cohorts, retention, adoption, and interpretation across product stages. | Wishlist |
| 20 | `product-strategist` | `alirezarezvani/claude-skills`, `product-team/skills/product-strategist/SKILL.md` | 23,170 | Product strategy | Test OKR cascades, strategic templates, alignment scoring, competitive analysis, and team design. | Wishlist |
| 21 | `notion-knowledge-capture` | `openai/skills`, `skills/.curated/notion-knowledge-capture/SKILL.md` | 24,179 | Knowledge base, Notion capture | Turn raw notes into structured, linked, reusable Notion pages. | Wishlist |
| 22 | `notion-meeting-intelligence` | `openai/skills`, `skills/.curated/notion-meeting-intelligence/SKILL.md` | 24,179 | Knowledge base, meeting memory | Connect meeting decisions, owners, actions, and prior context. | Wishlist |
| 23 | `notion-research-documentation` | `openai/skills`, `skills/.curated/notion-research-documentation/SKILL.md` | 24,179 | Knowledge base, research | Retrieve Notion sources and publish a cited, traceable synthesis. | Wishlist |
| 24 | `obsidian-markdown` | `kepano/obsidian-skills`, `skills/obsidian-markdown/SKILL.md` | 43,309 | Knowledge base, Obsidian | Build a connected vault with properties, wikilinks, embeds, callouts, and maps of content. | Wishlist |
| 25 | `obsidian-bases` | `kepano/obsidian-skills`, `skills/obsidian-bases/SKILL.md` | 43,309 | Knowledge base, structured views | Create database-like knowledge views with filters, formulas, layouts, and summaries. | Wishlist |
| 26 | `docx` | `anthropics/skills`, `skills/docx/SKILL.md` | 164,196 | Office, Word documents | Produce a styled Word report with tables, pagination, and rendered QA. | Wishlist |
| 27 | `pdf` | `anthropics/skills`, `skills/pdf/SKILL.md` | 164,196 | Office, PDF | Assemble and validate an accessible, client-ready PDF page by page. | Wishlist |
| 28 | `pptx` | `anthropics/skills`, `skills/pptx/SKILL.md` | 164,196 | Office, presentations | Build an editable executive presentation with narrative and slide-level QA. | Wishlist |
| 29 | `doc-coauthoring` | `anthropics/skills`, `skills/doc-coauthoring/SKILL.md` | 164,196 | Office, collaborative writing | Coauthor a decision document through context transfer, iteration, and reader testing. | Wishlist |
| 30 | `internal-comms` | `anthropics/skills`, `skills/internal-comms/SKILL.md` | 164,196 | Office, internal communication | Create a concise internal package spanning leadership notes, status, and FAQs. | Wishlist |
| 31 | `explore-data` | `anthropics/knowledge-work-plugins`, `data/skills/explore-data/SKILL.md` | 23,042 | Data to dashboard, exploration | Profile data around a decision and surface quality issues, patterns, and open questions. | Wishlist |
| 32 | `validate-data` | `anthropics/knowledge-work-plugins`, `data/skills/validate-data/SKILL.md` | 23,042 | Data to dashboard, validation | Establish grain, reconcile metrics, and expose quality checks before visualization. | Wishlist |
| 33 | `data-visualization` | `anthropics/knowledge-work-plugins`, `data/skills/data-visualization/SKILL.md` | 23,042 | Data to dashboard, encoding | Match analytical questions to visual encodings while preserving uncertainty. | Wishlist |
| 34 | `create-viz` | `anthropics/knowledge-work-plugins`, `data/skills/create-viz/SKILL.md` | 23,042 | Data to dashboard, chart creation | Produce accessible production charts with labels, tooltips, and narrative context. | Wishlist |
| 35 | `build-dashboard` | `anthropics/knowledge-work-plugins`, `data/skills/build-dashboard/SKILL.md` | 23,042 | Data to dashboard, product | Build an interactive decision dashboard with governed KPIs, filters, and action thresholds. | Wishlist |
| 36 | `marketing-plan` | `coreyhaines31/marketingskills`, `skills/marketing-plan/SKILL.md` | 41,714 | Marketing, strategy | Connect positioning, audience, channels, campaigns, and measurement in one plan. | Wishlist |
| 37 | `customer-research` | `coreyhaines31/marketingskills`, `skills/customer-research/SKILL.md` | 43,591 | Marketing, customer insight | Built KOL Evidence Studio's source-linked signal desk, five-theme synthesis, quote bank, confidence system, bias guardrails, and research-gap map. | Published |
| 38 | `content-strategy` | `coreyhaines31/marketingskills`, `skills/content-strategy/SKILL.md` | 43,591 | Marketing, content | Built KOL Content Engine's four-pillar strategy, hub-and-spoke map, buyer-stage backlog, and weighted priority filter. | Published |
| 39 | `copywriting` | `coreyhaines31/marketingskills`, `skills/copywriting/SKILL.md` | 43,591 | Marketing, conversion copy | Built a complete proof-led landing page with one action, three headline directions, benefits, process, comparison, objections, FAQ, and copy annotations. | Published |
| 40 | `analytics` | `coreyhaines31/marketingskills`, `skills/analytics/SKILL.md` | 43,591 | Marketing, measurement | Built a six-event contract, consent-aware local instrumentation, UTM context, live event inspector, QA checks, funnel, and attribution limits. | Published |
| 41 | `business-pulse` | `anthropics/knowledge-work-plugins`, `small-business/skills/business-pulse/SKILL.md` | 23,042 | One-person company, operations | Build a weekly founder cockpit for cash, pipeline, customers, delivery, and decisions. | Wishlist |
| 42 | `cash-flow-snapshot` | `anthropics/knowledge-work-plugins`, `small-business/skills/cash-flow-snapshot/SKILL.md` | 23,042 | One-person company, cash flow | Make runway, obligations, scenarios, and action triggers visible. | Wishlist |
| 43 | `lead-triage` | `anthropics/knowledge-work-plugins`, `small-business/skills/lead-triage/SKILL.md` | 23,042 | One-person company, sales | Score lead fit, urgency, next action, and follow-up timing. | Wishlist |
| 44 | `invoice-chase` | `anthropics/knowledge-work-plugins`, `small-business/skills/invoice-chase/SKILL.md` | 23,042 | One-person company, receivables | Create respectful follow-up stages with aging, exceptions, and approval gates. | Wishlist |
| 45 | `contract-review` | `anthropics/knowledge-work-plugins`, `small-business/skills/contract-review/SKILL.md` | 23,042 | One-person company, contracts | Flag obligations, risky terms, questions, and points for human escalation. | Wishlist |
| 46 | `brandkit` | `nexu-io/open-design`, `skills/brandkit/SKILL.md` | 81,644 | Design, brand systems | Build a reusable brand kit covering principles, tokens, typography, color, imagery, and applications. | Wishlist |
| 47 | `creative-director` | `nexu-io/open-design`, `skills/creative-director/SKILL.md` | 81,644 | Design, creative direction | Turn one strategic concept into a coherent visual campaign system across several formats. | Wishlist |
| 48 | `apple-hig` | `nexu-io/open-design`, `skills/apple-hig/SKILL.md` | 81,644 | Product design, Apple platforms | Audit a native-feeling product flow for hierarchy, navigation, controls, accessibility, and platform conventions. | Wishlist |
| 49 | `color-expert` | `nexu-io/open-design`, `skills/color-expert/SKILL.md` | 81,644 | Design, color systems | Test a semantic color system across themes, states, contrast requirements, data, and brand expression. | Wishlist |
| 50 | `design-brief` | `nexu-io/open-design`, `skills/design-brief/SKILL.md` | 81,644 | Design, briefing | Create a decision-ready brief connecting audience, problem, promise, constraints, deliverables, and success. | Wishlist |
| 51 | `embedding-strategies` | `wshobson/agents`, `plugins/llm-application-dev/skills/embedding-strategies/SKILL.md` | 38,241 | LLMs, embeddings | Benchmark models, chunking, similarity behavior, cost, latency, and retrieval quality. | Wishlist |
| 52 | `hybrid-search-implementation` | `wshobson/agents`, `plugins/llm-application-dev/skills/hybrid-search-implementation/SKILL.md` | 38,241 | LLMs, hybrid retrieval | Compare keyword, vector, and fused ranking with traceable relevance judgments. | Wishlist |
| 53 | `prompt-engineering-patterns` | `wshobson/agents`, `plugins/llm-application-dev/skills/prompt-engineering-patterns/SKILL.md` | 38,241 | LLMs, prompting | Build a pattern lab with fixed test cases, structured outputs, failure analysis, and regressions. | Wishlist |
| 54 | `vector-index-tuning` | `wshobson/agents`, `plugins/llm-application-dev/skills/vector-index-tuning/SKILL.md` | 38,241 | LLMs, vector indexes | Balance recall, latency, memory, filtering, and updates on a bounded corpus. | Wishlist |
| 55 | `multi-reviewer-patterns` | `wshobson/agents`, `plugins/agent-teams/skills/multi-reviewer-patterns/SKILL.md` | 38,241 | AI agents, review teams | Test distinct reviewer roles, severity rules, disagreement handling, and a measurable release decision. | Wishlist |
| 56 | `dataset-curation` | `wshobson/agents`, `plugins/llm-finetuning/skills/dataset-curation/SKILL.md` | 38,241 | ML, training data | Build a dataset with provenance, deduplication, quality labels, splits, contamination checks, and error analysis. | Wishlist |
| 57 | `finetuning-method-selection` | `wshobson/agents`, `plugins/llm-finetuning/skills/finetuning-method-selection/SKILL.md` | 38,241 | ML, fine-tuning strategy | Compare prompting, retrieval, adapters, and full tuning across quality, cost, data, and risk. | Wishlist |
| 58 | `lora-qlora-recipes` | `wshobson/agents`, `plugins/llm-finetuning/skills/lora-qlora-recipes/SKILL.md` | 38,241 | ML, LoRA and QLoRA | Run an adapter-tuning experiment with baseline evaluation, memory tracking, checkpoints, and failure cases. | Wishlist |
| 59 | `ml-pipeline-workflow` | `wshobson/agents`, `plugins/machine-learning-ops/skills/ml-pipeline-workflow/SKILL.md` | 38,241 | ML, MLOps | Build reproducible features, training, evaluation, registry, deployment checks, and drift monitoring. | Wishlist |
| 60 | `recsys-pipeline-architect` | `wshobson/agents`, `plugins/machine-learning-ops/skills/recsys-pipeline-architect/SKILL.md` | 38,241 | ML, recommender systems | Connect candidate generation, ranking, metrics, bias checks, serving constraints, and feedback loops. | Wishlist |
| 61 | `data-quality-frameworks` | `wshobson/agents`, `plugins/data-engineering/skills/data-quality-frameworks/SKILL.md` | 38,241 | Data engineering, quality | Define contracts, freshness, completeness, validity, reconciliation, ownership, and incident thresholds. | Wishlist |
| 62 | `dbt-transformation-patterns` | `wshobson/agents`, `plugins/data-engineering/skills/dbt-transformation-patterns/SKILL.md` | 38,467 | Data engineering, dbt | Built a layered dbt project with source freshness, staging and marts, documentation, lineage, incremental merge logic, tests, and failure simulation. | Published |
| 63 | `airflow-dag-patterns` | `wshobson/agents`, `plugins/data-engineering/skills/airflow-dag-patterns/SKILL.md` | 38,467 | Data engineering, orchestration | Built an idempotent TaskFlow DAG with retries, backfills, observability, quality branching, quarantine, cleanup, and recovery simulation. | Published |
| 64 | `competitive-landscape` | `wshobson/agents`, `plugins/startup-business-analyst/skills/competitive-landscape/SKILL.md` | 38,241 | Product, competitive strategy | Build an evidence-backed landscape of alternatives, segments, positioning, gaps, and implications. | Wishlist |
| 65 | `startup-metrics-framework` | `wshobson/agents`, `plugins/startup-business-analyst/skills/startup-metrics-framework/SKILL.md` | 38,241 | Product, startup metrics | Link acquisition, activation, retention, revenue, efficiency, and decision thresholds by stage. | Wishlist |
| 66 | `copy-editing` | `coreyhaines31/marketingskills`, `skills/copy-editing/SKILL.md` | 41,754 | Content writing, editing | Improve clarity, structure, voice, evidence, consistency, and conversion without changing meaning. | Wishlist |
| 67 | `cold-email` | `coreyhaines31/marketingskills`, `skills/cold-email/SKILL.md` | 41,754 | Content writing, outreach | Build a researched, respectful outreach sequence with relevance, proof, follow-ups, and reply metrics. | Wishlist |
| 68 | `emails` | `coreyhaines31/marketingskills`, `skills/emails/SKILL.md` | 44,208 | Content writing, email sequences | Built a five-email, ten-day learning journey with one job and CTA per message, role segments, entry, pause, and exit logic. | Published |
| 69 | `lead-magnets` | `coreyhaines31/marketingskills`, `skills/lead-magnets/SKILL.md` | 44,208 | Content writing, guides | Built a printable 16-point evaluation checklist with an ungated preview, optional capture, distribution, and measurement plan. | Published |
| 70 | `public-relations` | `coreyhaines31/marketingskills`, `skills/public-relations/SKILL.md` | 41,754 | Content writing, PR | Assemble a media-ready narrative with a defensible angle, supporting facts, spokesperson notes, and outreach. | Wishlist |
| 71 | `social` | `coreyhaines31/marketingskills`, `skills/social/SKILL.md` | 43,591 | Content writing, social | Built five native content atoms, a weekly cadence, engagement routine, short-video script, and evidence-first learning loop. | Published |
| 72 | `programmatic-seo` | `coreyhaines31/marketingskills`, `skills/programmatic-seo/SKILL.md` | 41,754 | Content writing, programmatic SEO | Test a useful page template, differentiated data, internal links, duplicate safeguards, and indexation quality. | Wishlist |
| 73 | `ai-seo` | `coreyhaines31/marketingskills`, `skills/ai-seo/SKILL.md` | 41,754 | Content writing, AI search | Build a source-rich answer library with clear entities, quotable facts, human usefulness, and attribution monitoring. | Wishlist |
| 74 | `offers` | `coreyhaines31/marketingskills`, `skills/offers/SKILL.md` | 41,754 | Content writing, offers | Connect outcome, audience, scope, evidence, risk reversal, pricing logic, objections, and a next step. | Wishlist |
| 75 | `product-marketing` | `coreyhaines31/marketingskills`, `skills/product-marketing/SKILL.md` | 43,591 | Content writing, product messaging | Built a canonical context and Positioning Room covering audience, problems, alternatives, differentiation, switching forces, language, voice, proof, and goals. | Published |
| 76 | `notion-spec-to-implementation` | `openai/skills`, `skills/.curated/notion-spec-to-implementation/SKILL.md` | 24,191 | Knowledge base, delivery | Turn a Notion specification into linked tasks, decisions, progress evidence, and durable project memory. | Wishlist |
| 77 | `defuddle` | `kepano/obsidian-skills`, `skills/defuddle/SKILL.md` | 43,329 | Knowledge base, web capture | Preserve useful web content, structure, metadata, citations, and readable Markdown. | Wishlist |
| 78 | `json-canvas` | `kepano/obsidian-skills`, `skills/json-canvas/SKILL.md` | 43,329 | Knowledge base, visual maps | Connect notes, sources, claims, questions, and decisions through inspectable nodes and edges. | Wishlist |
| 79 | `obsidian-cli` | `kepano/obsidian-skills`, `skills/obsidian-cli/SKILL.md` | 43,329 | Knowledge base, automation | Safely automate vault search, capture, metadata updates, link checks, and batch maintenance. | Wishlist |
| 80 | `library-curator` | `nexu-io/open-design`, `skills/library-curator/SKILL.md` | 81,644 | Knowledge base, curation | Build a resource library with inclusion rules, taxonomy, provenance, duplicate handling, and quality signals. | Wishlist |
| 81 | `weread-year-in-review-video-template` | `nexu-io/open-design`, `skills/weread-year-in-review-video-template/SKILL.md` | 81,644 | Knowledge base, WeRead | Turn books, highlights, themes, reading patterns, and memorable ideas into an editable year-in-review story. | Wishlist |
| 82 | `cdutcm-library-search` | `yoqalux-blip/cdutcm-library-search-skill`, `SKILL.md` | 0 | Chinese medicine KB, library research | Research-only authorized library search with source queues, citations, manual verification, and no medical advice. | Wishlist |
| 83 | `tcm-network-pharmacology` | `YuLab-SMU/TCMDATA`, `inst/skills/tcm-network-pharmacology/SKILL.md` | 15 | Chinese medicine KB, network pharmacology | Build a research evidence map that separates predictions from literature and experimental evidence; no medical advice. | Wishlist |
| 84 | `linglanmidian-systematic-evaluation-tcm` | `ndpvt-web/arxiv-claude-skills`, `skills/linglanmidian-systematic-evaluation-tcm/SKILL.md` | 6 | Chinese medicine KB, LLM evaluation | Create a source-grounded TCM knowledge benchmark with synonym-tolerant scoring and expert review. | Wishlist |
| 85 | `domain-ontology-mapping-tcm-and-ethnobotany` | `HolobiomicsLab/asb-skill-collections`, `collections/metabolomics/v2/skills/domain-ontology-mapping-tcm-and-ethnobotany/SKILL.md` | 14 | Chinese medicine KB, ontology mapping | Reconcile TCM names, transliterations, organisms, preparations, structures, and references with expert and license review. | Wishlist |
| 86 | `scikit-learn` | `K-Dense-AI/scientific-agent-skills`, `skills/scikit-learn/SKILL.md` | 33,024 | ML, classical modeling | Build a leakage-safe tabular ML benchmark with preprocessing pipelines, cross-validation, tuning, calibration, diagnostics, and a model card. | Wishlist |
| 87 | `shap` | `K-Dense-AI/scientific-agent-skills`, `skills/shap/SKILL.md` | 33,024 | ML, explainability, audit | Compare global, cohort, and local explanations with additivity checks, careful background data, and explicit non-causal limits. | Wishlist |
| 88 | `polars` | `K-Dense-AI/scientific-agent-skills`, `skills/polars/SKILL.md` | 33,024 | Data engineering, analytics | Benchmark a pandas-to-Polars migration using expressions, lazy plans, streaming, joins, schema controls, runtime, and memory evidence. | Wishlist |
| 89 | `timesfm-forecasting` | `K-Dense-AI/scientific-agent-skills`, `skills/timesfm-forecasting/SKILL.md` | 33,024 | ML, time-series forecasting | Compare zero-shot TimesFM forecasts with seasonal-naive baselines across horizons, prediction intervals, system limits, and failure cases. | Wishlist |
| 90 | `pytorch-lightning` | `K-Dense-AI/scientific-agent-skills`, `skills/pytorch-lightning/SKILL.md` | 33,024 | ML, scalable deep learning | Build a reproducible training system with modules, data loaders, callbacks, logging, checkpoints, distributed choices, and recovery QA. | Wishlist |
| 91 | `torch-geometric` | `K-Dense-AI/scientific-agent-skills`, `skills/torch-geometric/SKILL.md` | 33,024 | ML, graph neural networks | Compare GCN, GAT, and GraphSAGE on a bounded graph task with sampling, leakage checks, explainability, and scale constraints. | Wishlist |
| 92 | `aeon` | `K-Dense-AI/scientific-agent-skills`, `skills/aeon/SKILL.md` | 33,024 | ML, time-series learning | Test forecasting, classification, anomaly detection, segmentation, and similarity search with time-aware evaluation. | Wishlist |
| 93 | `pymc` | `K-Dense-AI/scientific-agent-skills`, `skills/pymc/SKILL.md` | 33,024 | Data science, Bayesian modeling | Build a hierarchical decision model with prior predictive checks, NUTS diagnostics, posterior checks, LOO comparison, and uncertainty-led decisions. | Wishlist |
| 94 | `statistical-analysis` | `K-Dense-AI/scientific-agent-skills`, `skills/statistical-analysis/SKILL.md` | 33,024 | Statistics, research analysis | Connect an estimand to test selection, assumptions, effect sizes, uncertainty, multiple comparisons, and reproducible reporting. | Wishlist |
| 95 | `statistical-power` | `K-Dense-AI/scientific-agent-skills`, `skills/statistical-power/SKILL.md` | 33,024 | Statistics, study planning | Build a sample-size planner using SESOI, closed-form and simulation power, clustering and dropout adjustments, and sensitivity curves. | Wishlist |
| 96 | `seaborn` | `K-Dense-AI/scientific-agent-skills`, `skills/seaborn/SKILL.md` | 33,024 | Data visualization, EDA | Create an exploratory atlas spanning distributions, relationships, categorical comparisons, facets, confidence intervals, and accessible themes. | Wishlist |
| 97 | `matplotlib` | `K-Dense-AI/scientific-agent-skills`, `skills/matplotlib/SKILL.md` | 33,024 | Data visualization, publication graphics | Produce a precisely styled multi-panel figure with custom artists, annotations, accessible encodings, export QA, and figure metadata. | Wishlist |
| 98 | `geopandas` | `K-Dense-AI/scientific-agent-skills`, `skills/geopandas/SKILL.md` | 33,024 | Data science, vector geospatial | Audit local vector data for CRS, geometry validity, spatial joins, area and distance caveats, provenance, and truthful mapping. | Wishlist |
| 99 | `geomaster` | `K-Dense-AI/scientific-agent-skills`, `skills/geomaster/SKILL.md` | 33,024 | Data science, geospatial ML | Combine STAC discovery, raster and vector processing, spatial features, model evaluation, and map-based uncertainty. | Wishlist |
| 100 | `networkx` | `K-Dense-AI/scientific-agent-skills`, `skills/networkx/SKILL.md` | 33,024 | Data science, network analysis | Test graph construction, components, paths, centrality, communities, robustness, file exchange, and explainable layouts. | Wishlist |
| 101 | `umap-learn` | `K-Dense-AI/scientific-agent-skills`, `skills/umap-learn/SKILL.md` | 33,024 | ML, dimensionality reduction | Compare settings, seeds, neighborhood preservation, supervised variants, density views, and downstream clustering stability. | Wishlist |
| 102 | `dask` | `K-Dense-AI/scientific-agent-skills`, `skills/dask/SKILL.md` | 33,024 | Data engineering, distributed analytics | Test partitioning, lazy execution, task graphs, cluster diagnostics, persistence choices, and pandas parity beyond memory. | Wishlist |
| 103 | `vaex` | `K-Dense-AI/scientific-agent-skills`, `skills/vaex/SKILL.md` | 33,024 | Data engineering, out-of-core analytics | Prototype memory-mapped, lazy analysis with virtual columns, fast aggregations, density visualization, and ML transforms. | Wishlist |
| 104 | `stable-baselines3` | `K-Dense-AI/scientific-agent-skills`, `skills/stable-baselines3/SKILL.md` | 33,024 | ML, reinforcement learning | Run a bounded RL experiment with environment checks, algorithm selection, seeded training, callbacks, evaluation, checkpoints, and learning curves. | Wishlist |
| 105 | `chart-visualization` | `antvis/chart-visualization-skills`, `skills/chart-visualization/SKILL.md` | 455; about 5.2K Skills.sh installs | Data visualization, chart selection | Test chart selection and rendering across time, comparison, distribution, hierarchy, and flow data, including API privacy and failure handling. | Wishlist |

## Promotion Checklist

Before a Skill becomes a public recommendation, capture:

| Field | Required note |
| --- | --- |
| Core concepts | The key Skill functions, methods, or concepts learned from the source. |
| 80% concept coverage | How the test used about 80% of those key concepts, and what was skipped. |
| Ground-up artifact | What was built from scratch using the Skill. |
| Rating out of 10 | High-level field rating after the build. |
| 3 good points | What made the Skill genuinely useful in practice. |
| 3 can-be-better points | Honest improvement notes, without recommending risky Skills. |
| Daily use cases | How a data person, AI builder, C-suite reader, student, or automation learner could use it. |
| Demo evidence | Product, flowchart, mindmap, screenshot, writeup, or other artifact made while testing. |
| Source and stars | GitHub repo, reference path, current star snapshot, and snapshot date. |
| Recommendation decision | Published or rejected. |