Claude Code Skills · page 126

This skill should be used when working with annotated data matrices in Python, particularly for single-cell genomics analysis, managing experimental measurements with metadata, or handling large-scale biological datasets. Use when tasks involve AnnData objects, h5ad files, single-cell RNA-seq data, or integration with scanpy/scverse tools.

Browse the web, take screenshots, interact with pages, fill forms. Use when: researching online, checking websites, filling out web forms, taking screenshots.

Read, search, send, and manage emails via Gmail API. Use when: managing email on the user's behalf, drafting replies, searching for specific emails, organizing with labels.

View, create, update, and delete Google Calendar events. Use when: scheduling meetings, checking availability, managing calendar events.

Search, read, upload, and share files in Google Drive

Read, write, and manage Google Sheets spreadsheets

Search, read, create, and manage Notion pages and databases

Create and manage scheduled tasks with cron, interval, or one-time execution

Search messages, manage channels, reactions, pins, and users in Slack

Draft release notes from merged PRs, propose a semver bump, and emit a copy-pasteable `gh release create` command. Use when preparing a tagged GitHub release.

>-

Web accessibility patterns for news and academic sites. Use for WCAG audits, alt text, accessible data viz, and assistive tech.

Electron desktop apps with React, TypeScript, and Vite. Use for IPC, window/tray, PTY terminals, WebRTC, and packaging.

Remote JavaScript console and debugging on mobile. Use for phone/tablet console errors, responsive testing, Eruda, and vConsole.

Flags irreversible decisions before commit. Use for data models, infra, auth boundaries, API contracts, event schemas, CI/CD.

Python data pipelines with modular architecture. Use for content workflows, batch jobs, or Google Sheets/Drive integration.

Enforces a test-driven bug-fixing workflow. Use when a user reports a bug, failing code, an error, or asks to fix something.

AI-assisted coding with Claude Code, Cursor, Copilot, Codex, Aider, or Windsurf. Use to build or debug generated code.

Authorized web scraping with fallback cascades and access-failure handling. Use for social media, yt-dlp, CAPTCHA or 403 blocks.

Signs of taste in web UI. Use when building or reviewing web interfaces, dashboards, SaaS apps, or internal tools.

Zero-build frontend development for static apps, browser extensions, maps, and lightweight data-backed interfaces. Use when deployment must not require a build step.

Eliminates AI-generated writing patterns that erode reader trust. Use when writing or editing articles, docs, or press releases.

Crisis communication and rapid-response workflows. Use when covering breaking news or coordinating fast fact-checking on deadline.

Acquire, clean, analyze, verify, visualize, and explain data for journalism. Use for reproducible data reporting, statistical analysis, maps, or public methodology.

Manages newsroom editorial workflows. Use when tracking assignments, managing deadlines, or coordinating editorial calendars.

Structured fact-checking workflow. Use when verifying claims for publication, rating claims, or setting pre-publication checks.

FOIA and public records workflows. Use when drafting requests, tracking submissions, checking exemptions, or appealing denials.

Interview preparation and recording-consent law. Use for research, question frameworks, one-party or all-party consent rules.

Transcription, recording management, and quote extraction. Use when processing audio/video or generating timestamped transcripts.

Email newsletter workflows. Use when creating newsletters, building subscriber lists, designing templates, or tracking engagement.

Enforces AP Style and newsroom conventions. Use when writing news, editing drafts, creating headlines, or tightening copy.

Social media monitoring, narrative tracking, and OSINT. Use when tracking viral spread, coordinated campaigns, or account vetting.

Verify sources, claims, images, video, documents, interviews, and synthetic media with SIFT and a durable evidence trail. Use before relying on uncertain material.

Crafts story pitches for different outlets. Use when pitching editors, writing query letters, or shaping feature or op-ed angles.

Designs PDF reports and proposals from HTML with previews and branding. Use to create, export, or securely upload a PDF.

Creates print-ready HTML that exports to PDF. Use to make a proposal, report, one-pager, newsletter, slides, or flyer.

Generates CLAUDE.md project memory files that transfer institutional knowledge. Use when setting up projects or onboarding.

Generates LESSONS.md files capturing institutional knowledge and failures. Use when closing out projects or investigations.

Selects the correct CLAUDE.md or LESSONS.md template via decision trees. Use when starting a project or unsure which fits.

Scholarly writing and research compliance. Use for CRediT, preregistration, Plan S, Nelson Memo, preprints, ORCID, LLM disclosure.

Legal methods for paywalled and geo-blocked content. Use for paywalls, academic papers, or open access via Unpaywall.

Digital archiving with AI enrichment and entity extraction. Use when building content archives or knowledge graphs.

Curated free APIs with verified free-tier limits. Use when suggesting APIs or finding free weather, news, finance, or sports data.

Web page change detection, availability tracking, and RSS feed generation. Use to monitor changes, downtime, or make a feed.

Web archiving and retrieval via Wayback Machine and Archive.today. Use to preserve content, reach dead pages, or save evidence.

API hardening for Express, FastAPI, and serverless. Use for rate limiting, CORS, input validation, API keys, or OWASP API Top 10.

Secure authentication patterns (OWASP, NIST). Use for login, registration, password reset, sessions, JWT, OAuth, MFA, passkeys.

Pre-deployment security audit organized by OWASP Top 10. Use when reviewing code before shipping or going to production.

Install-time cooldowns for npm/bun plus a sandboxed pre-install scan for bypasses. Use for supply-chain attacks or npm security.

Explores intent and design before implementation. MUST use before creating or modifying any feature, component, or behavior.

Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies

Use when you have a written implementation plan to execute in a separate session with review checkpoints

Presents options for merge, PR, or cleanup. Use when work is complete, tests pass, and you must decide how to integrate.

Verifies review feedback with technical rigor, not blind agreement. Use when receiving review before implementing suggestions.

Use when completing tasks, implementing major features, or before merging to verify work meets requirements

Use when executing implementation plans with independent tasks in the current session

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes

Use when implementing any feature or bugfix, before writing implementation code

Creates isolated git worktrees with smart directory selection and safety checks. Use when starting isolated feature work.

Establishes how to find and use skills, requiring Skill tool invocation before any response. Use when starting any conversation.

Runs verification commands and confirms output before success claims. Use before claiming work done or committing or PRs.

Use when you have a spec or requirements for a multi-step task, before touching code

Use when creating new skills, editing existing skills, or verifying skills work before deployment

HTML explainers, diagrams, architecture, timelines, source maps, slide decks, comparison tables, recaps, plan and diff reviews.

Embeds photo IPTC/EXIF/XMP metadata, caption, credit, alt text, license, AI-source label, GPS stripping, and C2PA credentials.

Scaffold a new Open Knowledge Format (OKF) knowledge base — a tree of small markdown concept files with YAML frontmatter, a spec, and a validator. Use when the user wants to start an OKF atlas/wiki/knowledge base, structure docs as one-concept-per-file with provenance, or initialize OKF in a repo (optionally into its GitHub wiki).

Scans the session for lessons and workflows, then proposes scoped CLAUDE.md edits. Use for save this lesson or add to context.

Builds an Open Knowledge Format (OKF) knowledge base from existing docs, notes, or a repo. Use to scaffold an OKF wiki.

Same-epoch comparison of training runs across wandb, neptune, tensorboard, or mlflow. Aligns runs at the student's current step (never current-vs-final-of-baseline) and separates proxy metrics from downstream targets. Use when the user asks to compare runs, check if a run is improving, track lag against a baseline, rank experiments, or evaluate run-vs-run performance.

>

Evidence-before-action diagnosis of failing ML experiments. Probes the system before guessing causes, process list, dmesg, GPU stats, log scrollback, checkpoint state, then states a hypothesis as a hypothesis and runs a smoke before claiming a root cause. Use when the user asks why a run is failing, diverging, OOMing, hanging, slow, producing weird metrics, has crashed, or asks to debug, diagnose, troubleshoot, or investigate a training issue.

>

>

Pre-flight checklist for long-running ML training jobs covering config diff, run naming, path verification, monitoring setup, and restart-cleanup. Use when the user asks to launch, kick off, start, restart, or kill a training run, or mentions launching a multi-hour or multi-day GPU job (python train, accelerate launch, torchrun, deepspeed, sbatch, tmux training).

>

>

>

End-to-end paper reproduction from arxiv URL through smoke runs to replication experiments. Handles missing or partial official code, missing training scripts, missing hyperparameters, and private datasets via similar-public-dataset substitution. Use when the user asks to reproduce, implement, replicate, or re-run a paper from scratch, or pastes an arxiv URL with reproduction intent.

>

>

Manages attention and evidence in long agent sessions. Use for lost instructions, dropped evidence, or large multi-agent contexts.

Directs the current request through configured lower-tier agents. Use only for explicit /director or /dev-toolkit:director invocation.

Public records requests under Brazil's Access to Information Law (LAI). Use for Fala.BR filings, e-SIC, and CGU/CMRI appeals.

Aggregates transcript and frame data into an interactive web dashboard. Use for content, topic, or sentiment analysis.

Collects videos from public social accounts. Use to download or scrape from Twitter, TikTok, YouTube, Instagram, or Facebook.

Extracts and visually analyzes frames from video files. Use for frame extraction, vision analysis, on-screen text, or frame grids.

Batch Whisper transcription of video or audio (WAV, podcasts) with a re-runnable provenance record. Use to transcribe recordings.

Surfaces relevant instincts during work. Use when starting a task to check if any learned behaviors apply.

Maintains awareness across sessions. Spawns observer agent on start, loads context, notifies of evolution opportunities.

基于佛教经典文献，生成特定高僧大德的 AI 教学角色

Use when user asks to compare masters, compare schools, compare perspectives, 对比, 各宗怎么看, 不同宗派, 禅净之争, 性相之辩, 空有之争, or wants multiple masters to answer the same question. Triggers include "对比"、"比较"、"各宗"、"不同宗派怎么看"、"禅宗和净土"、"天台和华严"、"唯识和中观"、"空有之争"、"性相之辩"、"各位祖师"、"多个角度"、"compare"、"comparison" — invoke whenever user's question implicitly or explicitly seeks multi-tradition perspectives on a Buddhist topic.

Use when user asks about 南传佛教, 上座部, Theravada, 巴利经典, 正念 sati, 放下, 三法印, 四念处, 出入息念 anapanasati, 戒定慧, 毗婆舍那, 森林禅林派, 巴蓬寺, 阿姜查, 杜多行, 中道, or wants teaching in 阿姜查 Ajahn Chah's voice. Triggers include "阿姜查"、"Ajahn Chah"、"森林禅"、"上座部"、"南传"、"巴利"、"正念"、"放下"、"禅修方法"、"妄念太多"、"打坐坐不住"、"巴蓬寺"、"杜多行"、"心的训练" — invoke whenever user's question touches Theravada / Thai Forest / mindfulness practice or asks about Ajahn Chah, even without explicit request.

Use when user asks about 藏传, 噶当派, Kadam, 三士道, 菩提道灯论, Bodhipathapradīpa, 阿底峡, Atiśa, 金洲大师, 七因果, 自他相换, 菩提心, 依止善知识, 暇满, 业果, 噶当六论, 仲敦巴, 热振寺, 藏地后弘期, or wants teaching in 阿底峡尊者 Atiśa's voice. Triggers include "阿底峡"、"觉沃杰"、"Atisha"、"Jowo Je"、"菩提道灯"、"道灯论"、"三士道"、"七因果"、"自他相换"、"金洲"、"噶当"、"仲敦巴"、"热振寺"、"道次第之祖" — invoke whenever user's question touches Kadam / lamrim foundations / bodhicitta cultivation, even without explicit request.

Use when user asks about 南传, 上座部, Theravāda, 巴利, 清净道论, Visuddhimagga, 戒定慧, 四十种业处, kammaṭṭhāna, 十遍, kasiṇa, 七清净, 十六观智, 阿毗达摩, Abhidhamma, 觉音尊者, Buddhaghosa, 大寺派, Mahāvihāra, 缘起十二支, 三法印, or wants teaching in 觉音尊者 Buddhaghosa's voice. Triggers include "觉音"、"Buddhaghosa"、"清净道论"、"Visuddhimagga"、"戒定慧三学"、"四十业处"、"七清净"、"十六观智"、"阿毗达摩注释"、"尼柯耶注释"、"上座部论师"、"大寺派" — invoke whenever user's question touches Theravāda commentarial / Visuddhimagga / Abhidhamma exegesis, even without explicit request.

Use when user asks for a sequenced learning path within a Buddhist tradition — 学修次第, 先学什么, 从哪入门, 下一步读什么, curriculum, 学习计划, 路径推荐. Differs from /compare-masters (parallel opinion) and /master-debate (adversarial dialectic) by being 纵向 / 时序: stage-by-stage plan keyed on tradition × level (L0-L3) → foundation → intermediate → advanced + blind spots. Trigger is planning intent — "禅宗对比" goes to /compare-masters; "禅宗从哪开始学" goes here.

Use when user explicitly asks for an adversarial / multi-round dialectic between masters — 祖师辩论, 各执一词, 谁更对, debate, 应成 vs 顿悟, 顿渐之争. Differs from /compare-masters (parallel single-round) by being adversarial multi-round via fresh-subagent orchestration. Topics 空有 / 禅净 / 性相 / 戒律 vs 内观 — trigger is adversarial framing: "禅净比较" → compare; "禅净辩论 / 谁更究竟" → here.

Use when user asks about 华严宗, 法界缘起, 四法界, 事事无碍, 十玄门, 六相圆融, 金师子章, 一即一切, 因陀罗网, 华严经, 五教判, or wants teaching in 法藏大师 Fazang's voice. Triggers include phrases like "华严"、"法藏"、"贤首"、"法界"、"事事无碍"、"十玄"、"六相"、"金师子"、"一即一切"、"理事无碍"、"因陀罗网"、"别教一乘"、"五教"、"毗卢遮那"、"一真法界" — invoke whenever user's question touches Huayan doctrine, even without explicit request.

Use when user asks about 禅宗, 六祖, 坛经, 顿悟, 见性成佛, 直指人心, 不立文字, 自性, 本心, 无念, 无相, 无住, 般若, 定慧一体, 明心见性, 南宗禅, or wants teaching in 慧能大师 Huineng's voice. Triggers include phrases like "禅"、"慧能"、"六祖"、"坛经"、"顿悟"、"见性"、"本来面目"、"菩提本无树"、"风动幡动"、"本来无一物"、"自性"、"机锋"、"烦恼即菩提"、"不二"、"弘忍" — invoke whenever user's question touches Chan/Zen doctrine, even without explicit request.

Use when user asks about 中观, 三论宗, 般若, 空性, 中道, 八不, 缘起性空, 法华经, 金刚经, 维摩诘, 不二法门, 一佛乘, 大智度论, or wants teaching in 鸠摩罗什 Kumārajīva's voice. Triggers include phrases like "中观"、"三论"、"空"、"般若"、"中道"、"八不"、"缘起性空"、"法华"、"金刚经"、"维摩诘"、"不二"、"实相"、"一佛乘"、"鸠摩罗什"、"罗什"、"会三归一"、"火宅"、"方便"、"中论"、"大智度论"、"百论"、"十二门论" — invoke whenever user's question touches Madhyamaka/Prajñā/Lotus doctrine, even without explicit request.

Use when user asks about 南传, 上座部, 缅甸内观, Mahasi Method, 标记法, Noting Method, 腹部起伏, 毗婆舍那, vipassanā, 四念处, 七清净, 十六观智, 刹那定, 行舍智, 马哈希尊者, Mahasi Sayadaw, Mahasi Sasana Yeiktha, IMS, or wants teaching in 马哈希尊者 Mahāsi Sayādaw's voice. Triggers include "马哈希"、"Mahasi"、"Sayadaw"、"标记法"、"腹部起伏"、"缅甸内观"、"密集禅修"、"十六观智"、"刹那定"、"妄念太多" — invoke whenever user's question touches Burmese vipassanā / Mahasi noting method, even without explicit request.