[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M73BC9N4706BZNWVHYKVA57D":3,"skill-same:01M73BC9N4706BZNWVHYKVA57D":76},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":28,"commitSha":29,"skillDirectory":30,"forksCount":31,"repositoryHtmlUrl":32,"repositoryHomepageUrl":6,"ownerHtmlUrl":33,"ownerBlogUrl":34,"ownerDescription":6,"updatedAt":35,"install":36,"resources":49},"01M73BC9N4706BZNWVHYKVA57D","causal-inference",null,"Production-grade Bayesian causal inference with PyMC, CausalPy, and DoWhy. Enforces DAG-first thinking, mandatory user checkpoints for assumptions, design-specific refutation, and defensible reporting with causal language guardrails. Trigger on: causal inference, causal effect estimation, treatment effects, counterfactuals, difference-in-differences (DiD), synthetic control, regression discontinuity (RDD), interrupted time series (ITS), instrumental variables (IV), propensity scores, DAGs, causal graphs, confounders, backdoor criterion, do-calculus, interventional distributions, pm.do(), pm.observe(), CausalPy, DoWhy, mediation analysis, refutation, sensitivity analysis, parallel trends, placebo tests, or any question of the form \"does X cause Y\" or \"what is the effect of X on Y.\"","qhjqhj00\u002Fresearch-skills-pool","qhjqhj00","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F29684092?v=4","01M72AE0ZGXVM5C57N512DGQG8","Tommy Chien","01M72AE107ESJMHT7QC5SXX144",3,"MIT","https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002Fbrycewang-stanford--Awesome-Agent-Skills-for-Empirical-Research--skills__23-Learning-Bayesian-Statistics-baygent-skills__causal-inference\u002FSKILL.md",[],[],"2026-11-11 15:11:14",false,"---\nname: causal-inference\ndescription: >\n  Production-grade Bayesian causal inference with PyMC, CausalPy, and DoWhy. Enforces DAG-first\n  thinking, mandatory user checkpoints for assumptions, design-specific refutation, and defensible\n  reporting with causal language guardrails. Trigger on: causal inference, causal effect estimation,\n  treatment effects, counterfactuals, difference-in-differences (DiD), synthetic control, regression\n  discontinuity (RDD), interrupted time series (ITS), instrumental variables (IV), propensity scores,\n  DAGs, causal graphs, confounders, backdoor criterion, do-calculus, interventional distributions,\n  pm.do(), pm.observe(), CausalPy, DoWhy, mediation analysis, refutation, sensitivity analysis,\n  parallel trends, placebo tests, or any question of the form \"does X cause Y\" or \"what is the\n  effect of X on Y.\"\nlicense: MIT\nmetadata:\n  author: \"[Alexandre Andorra](https:\u002F\u002Falexandorra.github.io\u002F)\"\n  version: \"1.0\"\n---\n\n# Causal Inference\n\n## Dependencies\n\nThis skill requires the **bayesian-workflow** skill for all PyMC modeling steps (priors, sampling,\ndiagnostics, calibration, reporting).\n\nDetect it:\n\n```bash\nls ~\u002F.claude\u002Fskills\u002Fbayesian-workflow\u002FSKILL.md 2>\u002Fdev\u002Fnull || ls .claude\u002Fskills\u002Fbayesian-workflow\u002FSKILL.md 2>\u002Fdev\u002Fnull\n```\n\nIf not found, install it:\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002FLearning-Bayesian-Statistics\u002Fbaygent-skills.git \u002Ftmp\u002Fbaygent-skills\ncp -r \u002Ftmp\u002Fbaygent-skills\u002Fbayesian-workflow ~\u002F.claude\u002Fskills\u002F\n```\n\nFor all PyMC modeling steps (priors, sampling, diagnostics, calibration, reporting), follow the\nbayesian-workflow skill.\n\n## Workflow overview\n\nEvery causal analysis follows this sequence. Steps 1-4 are the thinking phase (no code). Steps 5-8\nare the doing phase. Think before you do.\n\n1. **Formulate the causal question** — Propose precise estimand (ATE, ATT, LATE, etc.). ⚠️ ASK USER TO CONFIRM.\n2. **Draw the DAG** — Propose causal graph with nodes, edges, and explicit non-edges. ⚠️ ASK USER TO CONFIRM. See [references\u002Fdags-and-identification.md](references\u002Fdags-and-identification.md)\n3. **Identify** — Determine identification strategy (backdoor, front-door, IV, RDD, DiD). ⚠️ ASK USER TO CONFIRM untestable assumptions. See [references\u002Fdags-and-identification.md](references\u002Fdags-and-identification.md)\n4. **Choose design** — Match problem to method using table below. ⚠️ ASK USER TO CONFIRM. See [references\u002Fquasi-experiments.md](references\u002Fquasi-experiments.md) or [references\u002Fstructural-models.md](references\u002Fstructural-models.md)\n5. **Estimate** — Build and fit the model. Delegate all PyMC mechanics to bayesian-workflow skill.\n6. **Refute** — MANDATORY. Run design-specific robustness checks. See [references\u002Frefutation.md](references\u002Frefutation.md)\n7. **Interpret** — Effect size + decision-relevant HDIs + probability of direction.\n8. **Report** — Generate causal analysis report. See [references\u002Freporting.md](references\u002Freporting.md)\n\n## Design selection guide\n\n| Design | Use when | Key assumption | Tool |\n|---|---|---|---|\n| DiD | Treatment at known time, control group available | Parallel trends | CausalPy |\n| Staggered DiD | Treatment rolls out at different times | Parallel trends per cohort | CausalPy |\n| Synthetic Control | Single treated unit, donor pool available | Weighted donors approximate counterfactual | CausalPy |\n| ITS | Time series, intervention at known time, no control | No confounding event at treatment time | CausalPy |\n| RDD | Treatment by threshold on running variable | No manipulation at threshold | CausalPy |\n| IV | Endogenous treatment, valid instrument | Exclusion restriction, relevance | CausalPy |\n| IPSW | Observational data, treatment modeled | No unmeasured confounders, positivity | CausalPy |\n| Structural (do\u002Fobserve) | Full causal theory, model mechanisms | Correct DAG specification | PyMC |\n| Counterfactual | \"What would Y have been if X differed?\" | Correct structural model | PyMC |\n\n## Critical rules\n\n- **No estimation without a confirmed DAG.** A causal graph is not optional decoration — it makes\n  assumptions explicit and determines the adjustment set. If the user resists, explain why the DAG\n  is non-negotiable before proceeding.\n- **No causal claims without refutation.** Every design has failure modes. Run at minimum one\n  design-specific robustness check (placebo test, sensitivity analysis, falsification test) before\n  reporting results. See [references\u002Frefutation.md](references\u002Frefutation.md).\n- **State assumptions before results.** Lead with what must be true for the estimate to be causal.\n  Bury the estimate after the assumptions, not before. This is not optional politeness — it prevents\n  misuse of results.\n- **Adapt HDIs to the decision context.** The bayesian-workflow skill's 94% HDI is a sensible\n  default; adapt it with explicit explanation when the decision stakes warrant it (e.g., 89% for\n  exploratory, 97% for high-stakes policy). Report multiple intervals when the decision threshold\n  matters.\n- **Downgrade causal language when warranted.** If identification assumptions are unverifiable or\n  refutation raises flags, soften claims: \"consistent with a causal effect\" not \"causes\", \"estimated\n  effect\" not \"true effect\". Flag uncertainty loudly in the report.\n- **Ask the user when domain knowledge is needed.** You cannot know whether an instrument is valid,\n  whether parallel trends holds, or whether a confounder exists without domain expertise. Ask\n  before assuming.\n- **Delegate PyMC mechanics to bayesian-workflow.** This skill handles causal structure and design.\n  The bayesian-workflow skill handles priors, sampling, diagnostics, calibration, and reporting\n  format. Don't duplicate those rules here.\n\n## Common gotchas\n\nThese are battle-tested lessons that save hours of debugging:\n\n- **CausalPy formula syntax uses `C()` for categoricals.** Passing a string column directly without\n  `C()` will silently produce wrong dummy coding. Always wrap categorical treatment and group\n  variables: `\"y ~ C(treatment) + C(group)\"`.\n- **DoWhy requires explicit `U` nodes for unobserved confounders.** Omitting them from the graph\n  will make DoWhy treat your model as fully identified when it isn't. Add latent nodes explicitly\n  and mark them as unobserved.\n- **CausalPy's PyMC models don't auto-store log-likelihood.** Same issue as bayesian-workflow:\n  nutpie silently drops it. Call `pm.compute_log_likelihood(idata, model=model)` after sampling if\n  you need it for model comparison.\n- **Parallel trends is untestable in the post-treatment period.** Pre-treatment trend tests are\n  necessary but not sufficient — passing them doesn't prove the assumption holds after treatment.\n  State this explicitly in every DiD report.\n- **Synthetic control requires the treated unit to lie within the convex hull of donors.** If the\n  treated unit is an outlier (highest GDP, largest city), no weighted combination of donors can\n  approximate its counterfactual. Check this before running — if violated, the design is invalid.\n- **DiD group variable must be dummy-coded (0\u002F1).** CausalPy rejects string labels like \"treatment\"\u002F\"control\". Use integers: 1 = treatment, 0 = control. Data also requires a `unit` column.\n- **SyntheticControl expects wide-format data.** Index = time, columns = unit names, values = outcome. If your data is long format, pivot first: `df.pivot(index=\"date\", columns=\"unit\", values=\"outcome\")`.\n\n## When things go wrong\n\n| Symptom | Likely cause | Fix |\n|---|---|---|\n| Refutation fails | Assumption violated | Diagnose which assumption, try alternative design or sensitivity bounds |\n| DiD effect at placebo time | Parallel trends violated | Try synthetic control or add group-specific time trends |\n| RDD: bunching at threshold | Manipulation of running variable | Design is invalid for this threshold — report and stop |\n| SC: poor pre-treatment fit | Donors don't span treated unit | Add donors, expand donor pool, or reconsider design |\n| DoWhy says \"not identifiable\" | Insufficient adjustment set | Revise DAG, add measured variables, or change design |\n| CausalPy formula error | Wrong formula syntax | Use `C()` for categoricals, check variable names match dataframe columns |\n","# Causal Inference\n\n## Dependencies\n\nThis skill requires the **bayesian-workflow** skill for all PyMC modeling steps (priors, sampling,\ndiagnostics, calibration, reporting).\n\nDetect it:\n\n```bash\nls ~\u002F.claude\u002Fskills\u002Fbayesian-workflow\u002FSKILL.md 2>\u002Fdev\u002Fnull || ls .claude\u002Fskills\u002Fbayesian-workflow\u002FSKILL.md 2>\u002Fdev\u002Fnull\n```\n\nIf not found, install it:\n\n```bash\ngit clone https:\u002F\u002Fgithub.com\u002FLearning-Bayesian-Statistics\u002Fbaygent-skills.git \u002Ftmp\u002Fbaygent-skills\ncp -r \u002Ftmp\u002Fbaygent-skills\u002Fbayesian-workflow ~\u002F.claude\u002Fskills\u002F\n```\n\nFor all PyMC modeling steps (priors, sampling, diagnostics, calibration, reporting), follow the\nbayesian-workflow skill.\n\n## Workflow overview\n\nEvery causal analysis follows this sequence. Steps 1-4 are the thinking phase (no code). Steps 5-8\nare the doing phase. Think before you do.\n\n1. **Formulate the causal question** — Propose precise estimand (ATE, ATT, LATE, etc.). ⚠️ ASK USER TO CONFIRM.\n2. **Draw the DAG** — Propose causal graph with nodes, edges, and explicit non-edges. ⚠️ ASK USER TO CONFIRM. See [references\u002Fdags-and-identification.md](references\u002Fdags-and-identification.md)\n3. **Identify** — Determine identification strategy (backdoor, front-door, IV, RDD, DiD). ⚠️ ASK USER TO CONFIRM untestable assumptions. See [references\u002Fdags-and-identification.md](references\u002Fdags-and-identification.md)\n4. **Choose design** — Match problem to method using table below. ⚠️ ASK USER TO CONFIRM. See [references\u002Fquasi-experiments.md](references\u002Fquasi-experiments.md) or [references\u002Fstructural-models.md](references\u002Fstructural-models.md)\n5. **Estimate** — Build and fit the model. Delegate all PyMC mechanics to bayesian-workflow skill.\n6. **Refute** — MANDATORY. Run design-specific robustness checks. See [references\u002Frefutation.md](references\u002Frefutation.md)\n7. **Interpret** — Effect size + decision-relevant HDIs + probability of direction.\n8. **Report** — Generate causal analysis report. See [references\u002Freporting.md](references\u002Freporting.md)\n\n## Design selection guide\n\n| Design | Use when | Key assumption | Tool |\n|---|---|---|---|\n| DiD | Treatment at known time, control group available | Parallel trends | CausalPy |\n| Staggered DiD | Treatment rolls out at different times | Parallel trends per cohort | CausalPy |\n| Synthetic Control | Single treated unit, donor pool available | Weighted donors approximate counterfactual | CausalPy |\n| ITS | Time series, intervention at known time, no control | No confounding event at treatment time | CausalPy |\n| RDD | Treatment by threshold on running variable | No manipulation at threshold | CausalPy |\n| IV | Endogenous treatment, valid instrument | Exclusion restriction, relevance | CausalPy |\n| IPSW | Observational data, treatment modeled | No unmeasured confounders, positivity | CausalPy |\n| Structural (do\u002Fobserve) | Full causal theory, model mechanisms | Correct DAG specification | PyMC |\n| Counterfactual | \"What would Y have been if X differed?\" | Correct structural model | PyMC |\n\n## Critical rules\n\n- **No estimation without a confirmed DAG.** A causal graph is not optional decoration — it makes\n  assumptions explicit and determines the adjustment set. If the user resists, explain why the DAG\n  is non-negotiable before proceeding.\n- **No causal claims without refutation.** Every design has failure modes. Run at minimum one\n  design-specific robustness check (placebo test, sensitivity analysis, falsification test) before\n  reporting results. See [references\u002Frefutation.md](references\u002Frefutation.md).\n- **State assumptions before results.** Lead with what must be true for the estimate to be causal.\n  Bury the estimate after the assumptions, not before. This is not optional politeness — it prevents\n  misuse of results.\n- **Adapt HDIs to the decision context.** The bayesian-workflow skill's 94% HDI is a sensible\n  default; adapt it with explicit explanation when the decision stakes warrant it (e.g., 89% for\n  exploratory, 97% for high-stakes policy). Report multiple intervals when the decision threshold\n  matters.\n- **Downgrade causal language when warranted.** If identification assumptions are unverifiable or\n  refutation raises flags, soften claims: \"consistent with a causal effect\" not \"causes\", \"estimated\n  effect\" not \"true effect\". Flag uncertainty loudly in the report.\n- **Ask the user when domain knowledge is needed.** You cannot know whether an instrument is valid,\n  whether parallel trends holds, or whether a confounder exists without domain expertise. Ask\n  before assuming.\n- **Delegate PyMC mechanics to bayesian-workflow.** This skill handles causal structure and design.\n  The bayesian-workflow skill handles priors, sampling, diagnostics, calibration, and reporting\n  format. Don't duplicate those rules here.\n\n## Common gotchas\n\nThese are battle-tested lessons that save hours of debugging:\n\n- **CausalPy formula syntax uses `C()` for categoricals.** Passing a string column directly without\n  `C()` will silently produce wrong dummy coding. Always wrap categorical treatment and group\n  variables: `\"y ~ C(treatment) + C(group)\"`.\n- **DoWhy requires explicit `U` nodes for unobserved confounders.** Omitting them from the graph\n  will make DoWhy treat your model as fully identified when it isn't. Add latent nodes explicitly\n  and mark them as unobserved.\n- **CausalPy's PyMC models don't auto-store log-likelihood.** Same issue as bayesian-workflow:\n  nutpie silently drops it. Call `pm.compute_log_likelihood(idata, model=model)` after sampling if\n  you need it for model comparison.\n- **Parallel trends is untestable in the post-treatment period.** Pre-treatment trend tests are\n  necessary but not sufficient — passing them doesn't prove the assumption holds after treatment.\n  State this explicitly in every DiD report.\n- **Synthetic control requires the treated unit to lie within the convex hull of donors.** If the\n  treated unit is an outlier (highest GDP, largest city), no weighted combination of donors can\n  approximate its counterfactual. Check this before running — if violated, the design is invalid.\n- **DiD group variable must be dummy-coded (0\u002F1).** CausalPy rejects string labels like \"treatment\"\u002F\"control\". Use integers: 1 = treatment, 0 = control. Data also requires a `unit` column.\n- **SyntheticControl expects wide-format data.** Index = time, columns = unit names, values = outcome. If your data is long format, pivot first: `df.pivot(index=\"date\", columns=\"unit\", values=\"outcome\")`.\n\n## When things go wrong\n\n| Symptom | Likely cause | Fix |\n|---|---|---|\n| Refutation fails | Assumption violated | Diagnose which assumption, try alternative design or sensitivity bounds |\n| DiD effect at placebo time | Parallel trends violated | Try synthetic control or add group-specific time trends |\n| RDD: bunching at threshold | Manipulation of running variable | Design is invalid for this threshold — report and stop |\n| SC: poor pre-treatment fit | Donors don't span treated unit | Add donors, expand donor pool, or reconsider design |\n| DoWhy says \"not identifiable\" | Insufficient adjustment set | Revise DAG, add measured variables, or change design |\n| CausalPy formula error | Wrong formula syntax | Use `C()` for categoricals, check variable names match dataframe columns |\n",{"name":5,"license":15,"metadata":24,"description":27},{"author":25,"version":26},"[Alexandre Andorra](https:\u002F\u002Falexandorra.github.io\u002F)","1.0","Production-grade Bayesian causal inference with PyMC, CausalPy, and DoWhy. Enforces DAG-first thinking, mandatory user checkpoints for assumptions, design-specific refutation, and defensible reporting with causal language guardrails. Trigger on: causal inference, causal effect estimation, treatment effects, counterfactuals, difference-in-differences (DiD), synthetic control, regression discontinuity (RDD), interrupted time series (ITS), instrumental variables (IV), propensity scores, DAGs, causal graphs, confounders, backdoor criterion, do-calculus, interventional distributions, pm.do(), pm.observe(), CausalPy, DoWhy, mediation analysis, refutation, sensitivity analysis, parallel trends, placebo tests, or any question of the form \"does X cause Y\" or \"what is the effect of X on Y.\"\n","master","c247302c836de6438d746b45670990eb71691166","skills_collected\u002Fbrycewang-stanford--Awesome-Agent-Skills-for-Empirical-Research--skills__23-Learning-Bayesian-Statistics-baygent-skills__causal-inference",1,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool","https:\u002F\u002Fgithub.com\u002Fqhjqhj00","chien.io","2026-05-11 15:07:59",{"source":8,"skillName":5,"commands":37,"prompt":47,"zipAvailable":48},[38,41,44],{"tool":39,"command":40},"npx","npx skills add qhjqhj00\u002Fresearch-skills-pool --skill causal-inference",{"tool":42,"command":43},"bunx","bunx skills add qhjqhj00\u002Fresearch-skills-pool --skill causal-inference",{"tool":45,"command":46},"pnpm","pnpm dlx skills add qhjqhj00\u002Fresearch-skills-pool --skill causal-inference","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002Fbrycewang-stanford--Awesome-Agent-Skills-for-Empirical-Research--skills__23-Learning-Bayesian-Statistics-baygent-skills__causal-inference\u002FSKILL.md",true,[50,56,60,64,68,72],{"path":51,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":54,"sourceHtmlUrl":55},"README.md","REFERENCE","参考文档",4528,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002Fbrycewang-stanford--Awesome-Agent-Skills-for-Empirical-Research--skills__23-Learning-Bayesian-Statistics-baygent-skills__causal-inference\u002FREADME.md",{"path":57,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":58,"sourceHtmlUrl":59},"references\u002Fdags-and-identification.md",8556,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002Fbrycewang-stanford--Awesome-Agent-Skills-for-Empirical-Research--skills__23-Learning-Bayesian-Statistics-baygent-skills__causal-inference\u002Freferences\u002Fdags-and-identification.md",{"path":61,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":62,"sourceHtmlUrl":63},"references\u002Fquasi-experiments.md",19268,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002Fbrycewang-stanford--Awesome-Agent-Skills-for-Empirical-Research--skills__23-Learning-Bayesian-Statistics-baygent-skills__causal-inference\u002Freferences\u002Fquasi-experiments.md",{"path":65,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":66,"sourceHtmlUrl":67},"references\u002Frefutation.md",22685,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002Fbrycewang-stanford--Awesome-Agent-Skills-for-Empirical-Research--skills__23-Learning-Bayesian-Statistics-baygent-skills__causal-inference\u002Freferences\u002Frefutation.md",{"path":69,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":70,"sourceHtmlUrl":71},"references\u002Freporting.md",11013,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002Fbrycewang-stanford--Awesome-Agent-Skills-for-Empirical-Research--skills__23-Learning-Bayesian-Statistics-baygent-skills__causal-inference\u002Freferences\u002Freporting.md",{"path":73,"resourceType":52,"resourceTypeDesc":53,"sizeBytes":74,"sourceHtmlUrl":75},"references\u002Fstructural-models.md",7650,"https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Fskills_collected\u002Fbrycewang-stanford--Awesome-Agent-Skills-for-Empirical-Research--skills__23-Learning-Bayesian-Statistics-baygent-skills__causal-inference\u002Freferences\u002Fstructural-models.md",[77,84,91,98,105,112],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M73JJB6EQPFPPVQK2Y9N67Q0","transformer-lens-interpretability","Provides guidance for mechanistic interpretability research using TransformerLens to inspect and manipulate transformer internals via HookPoints and activation caching. Use when reverse-engineering model algorithms, studying attention patterns, or performing activation patching experiments.","https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Ftop100-runs\u002F130-transformer-lens\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":19},"01M73JHTANRRWEE9HP98TPEHAY","wiki-vitepress","Packages generated wiki Markdown into a VitePress static site with dark theme, dark-mode Mermaid diagrams with click-to-zoom, and production build output. Use when the user wants to create a browsable website from generated wiki pages.","https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Ftop100-runs\u002F126-wiki-vitepress\u002FSKILL.md",[],[],{"skillNo":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"publishedAt":19},"01M73JHJBMVFW46F1Y8K3SBF43","phoenix-observability","Open-source AI observability platform for LLM tracing, evaluation, and monitoring. Use when debugging LLM applications with detailed traces, running evaluations on datasets, or monitoring production AI systems with real-time insights.","https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Ftop100-runs\u002F124-phoenix\u002FSKILL.md",[],[],{"skillNo":99,"name":100,"nameZh":6,"description":101,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":102,"categories":103,"keywords":104,"publishedAt":19},"01M73JHFPA3A2806PMP42ZN86Q","frontend-ui-dark-ts","Build dark-themed React applications using Tailwind CSS with custom theming, glassmorphism effects, and Framer Motion animations. Use when creating dashboards, admin panels, or data-rich interfaces with a refined dark aesthetic.","https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Ftop100-runs\u002F123-frontend-ui-dark-ts\u002FSKILL.md",[],[],{"skillNo":106,"name":107,"nameZh":6,"description":108,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":109,"categories":110,"keywords":111,"publishedAt":19},"01M73JFVBVK57NRJS49TT2VDCT","tensorboard","Visualize training metrics, debug models with histograms, compare experiments, visualize model graphs, and profile performance with TensorBoard - Google's ML visualization toolkit","https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Ftop100-runs\u002F113-tensorboard\u002FSKILL.md",[],[],{"skillNo":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":116,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":119,"publishedAt":19},"01M73JFF6PK6ZCN41C89J7HS2B","markdown-mermaid-writing","Comprehensive markdown and Mermaid diagram writing skill. Use when creating any scientific document, report, analysis, or visualization. Establishes text-based diagrams as the default documentation standard with full style guides (markdown + mermaid), 24 diagram type references, and 9 document templates.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fqhjqhj00\u002Fresearch-skills-pool\u002Fblob\u002Fmaster\u002Ftop100-runs\u002F110-markdown-mermaid-writing\u002FSKILL.md",[],[]]