Research

Embeds harmful requests as completed security research entries in a threat intelligence database, where the model generates summaries of already-documented…

Create and operate a project-scoped research wiki that turns a human seed into an evidence-grounded writing guideline or software implementation plan.

Use when gathering a bounded signal inventory memo for an open source repository so maintainers can dispatch follow-up investigation work

Creates shareable artifacts: plans and proposals that others will review and comment on, research and investigation write-ups, comparisons, status reports…

Cite-or-Block 架构的基石原子 skill。把报告里的 citation 锚点 (例如 [guideline:CSCO-2024-NSCLC:§5.5.2] / [pmid:12345678:abstract]) 解析为源文件片段,并核对事实声明里的关键词是否在引用源原文里出现。 Foundation…

Phase 4 Turn 1 of disease-market-sizing-orchestration.

Unified Python interface to 40+ bioinformatics services.

Frames harmful requests as analytical taxonomy development for harm prevention research, avoiding generative examples.

Baidu Scholar Search - Search Chinese and English academic literature (journals, conferences, papers, etc.)

Review recent code changes, check if they are ready to ship, or assess a research brief.

本技能用于对中国大陆企业开展法律尽职调查，以及起草、生成、修改、完善、审查或复核法律尽职调查报告；支持投融资、并购、常年法律顾问合规体检、供应商或合作方、融资方、上市前专项及其他专项尽调。 即使用户未准确说出完整报告名称，只要任务实质是企业法律尽调，仍应优先使用本技能，不得由…

中国大陆非诉法律意见书生成与复核专用技能。用户明确要求“法律意见书”“专项法律意见书”“咨询回复式法律意见书”“出具法律意见”“形成正式法律意见”“法律意见书初稿”，或者要求起草、修改、完善、重构、审查、复核已有法律意见书时，必须使用本技能。 最终交付物只要是供律师、法务、管理层或项目决策使用的法律意见书…

Search arXiv papers by keyword, author, category or ID and read abstracts/PDFs (http_fetch on the arXiv API).

協助撰寫、縮短、驗證、整理可交給下一個 agent session（Claude Code 或 Codex）執行的 handoff prompt，含 Codex app 的 `/goal`。當使用者要寫 `/goal`、準備 handoff prompt、把模糊任務整理成可執行 goal、要依任務性質與剩餘…

Cite-or-Block 架构下的药名 cross-check 原子 skill。给定报告文本与 .cache/<slug>/sources/,扫描每个药品提及 → 找最近 citation 锚点 → 核对锚点指向的源里是否真的提到了此药(通用名 + 商品名)。无字典、无 fallback、源里没 =…

End-of-workflow audit of touched GitHub issues, PRs, and branches via a Sonnet subagent, plus a deterministic sweep of stale build-tool temp artifacts (e.g.

Searches the web and fetches pages through DeepSeek's Search and Fetch APIs (/v1/tools/search, /v1/tools/fetch) using DEEPSEEK_API_KEY, and routes model-side…

Uses Firecrawl's scrape, search, map, crawl, batch, parse, interact, and monitor endpoints at the lowest credit cost, never /agent or /extract.

Grounded Google Search, URL reading, and gated Deep Research via Gemini (GEMINI_API_KEY).

Searches the web through MiniMax's Token Plan search API (api.minimax.io/v1/coding_plan/search) using MINIMAX_API_KEY.

Uses Parallel (parallel.ai) Search, Extract, Task, Task Group, FindAll, Entity Search, Responses, and Monitor APIs at the lowest sensible cost.

Audit real codebases for SEO: metadata, rendering, crawlability, canonicals, schema, OG/social previews, i18n, sitemaps, robots, internal links, and AI-era…

Searches the web through Synthetic's zero-data-retention /v2/search API using SYNTHETIC_API_KEY.

Searches the web, reads pages, and searches GitHub repositories through Z.ai (Zhipu GLM) tools using Z_AI_API_KEY: web search, web reader, and Zread over MCP…