Намерете инструкция, прочетете оригиналния файл и подгответе команда за своя инструмент.
31 301 skills в каталога
Как подбирам и измервамДанните имат нужда от нова проверка.Последна успешна проверка: 23.09.2026 г.
Проверката на източника не означава, че всяка инструкция е изпробвана от Level8. Каталогът не обхваща непременно всички публични хранилища.
Търсене в целия индекс. Метаданните показват посоченото от източника; липсващите стойности остават означени.
Датата в GitHub е отделна от първата поява в моя индекс. Филтърът по дата изключва записите без посочено обновяване.
Страница 200: 24 записа в 15 групи по репозиторий. Всички записи от страницата са показани. Номерът пази позицията от избраната подредба.
2 skills от този източник на тази страница
Motivator & coach tingkat EXPERT berbasis data — bukan kata penyemangat kosong. Memakai kerangka profesional: model coaching GROW, Self-Determination Theory (autonomy–competence–relatedness), Progress Principle (Amabile), motivational interviewing (OARS), serta desain micro-wins dan ritual tim. Mena
Bekerja dengan modalitas selain teks: gambar (audit aset, rasio, metadata, prompt reverse-engineering), audio/TTS (penyiapan naskah & pacing), video (cut-list dari storyboard), dan dokumen (ekstrak tabel ke data). Menyediakan checklist QC per modalitas dan template prompt multimodal. Gunakan saat pe
4 skills от този източник на тази страница
Data-analysis report format: dataset provenance, methods, statistical results with effect sizes and CIs, figure/table references, and honest limitation notes. Use for 数据分析报告/统计结果报告. Pairs with the analysis runner's outputs (CSV/PNG/provenance).
Pythonic wrapper around RDKit with simplified interface and sensible defaults. Preferred for standard drug discovery including SMILES parsing, standardization, descriptors, fingerprints, clustering, 3D conformers, parallel processing. Returns native rdkit.Chem.Mol objects. For advanced control or custom parameters, use rdkit directly.
Molecular ML with diverse featurizers and pre-built datasets. Use for property prediction (ADMET, toxicity) with traditional ML or GNNs when you want extensive featurization options and MoleculeNet benchmarks. Best for quick experiments with pre-trained models, diverse molecular representations. For graph-first PyTorch workflows use torchdrug; for benchmark datasets use pytdc.
Perform comprehensive exploratory data analysis on scientific data files across 200+ file formats. This skill should be used when analyzing any scientific data file to understand its structure, content, quality, and characteristics. Automatically detects file type and generates detailed markdown reports with format-specific analysis, quality metrics, and downstream analysis recommendations. Covers chemistry, bioinformatics, microscopy, spectroscopy, proteomics, metabolomics, and general scientif
1 skill от този източник на тази страница
Interpret metrics or experiment results by separating fact from explanation and ruling out alternative causes before recommending action. Use when reviewing results and need to draw a real conclusion, not just eyeball a chart.
1 skill от този източник на тази страница
Normalize messy marketplace listings and score flip profit, repair cost, fees, and risk. Use when changing analyzer.py, deal_scorer.py, LLM/vision analysis, comps, or profit math.
1 skill от този източник на тази страница
Use when the user requests a research report, literature review, market or industry analysis, competitive landscape, due diligence, policy brief, or technical deep-dive — anything needing evidence from many web sources synthesized with citations. Also triggers on: "research this topic", "write a rep
1 skill от този източник на тази страница
Use when a desktop app's bundled agent must run on a remote host.
1 skill от този източник на тази страница
Use when orchestrating build, fix, spec, refactor, or harden workflows on kanban cards. Manages the full agent delegation loop with circuit breakers, branches, commits, and user handoffs.
2 skills от този източник на тази страница
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available
1 skill от този източник на тази страница
懂你(DongNi)用户偏好记忆——记住并跨会话应用用户的输出格式、语言风格、技术栈、沟通习惯等偏好。首次对话自动冷启动（3道选择题）；此后每次对话静默学习并更新画像文件 ~/.dongni/profile.json，让Agent自动适配用户习惯、消除重复交代。凡涉及"偏好/习惯/格式/风格/记住我/修改偏好"的请求，或用户输入 /dongni 时使用本Skill。
1 skill от този източник на тази страница
Maak van de agent een expert in evidence-based educatief presentatieontwerp. Gebruik bij het ontwerpen van presentaties voor onderwijs, training of leercontexten waarin retentie en begrip belangrijk zijn. Past Cognitive Load Theory, Mayers 12 principes voor multimediaal leren, Gagné's 9 instructiege
1 skill от този източник на тази страница
English text quality. Triggers: edit this, proofread, tighten this up, make this sound human, does this read like AI, en-text. Catches AI slop, buried subjects, nominalizations, and hedging. Applies on request; light cleanup on any English draft the agent writes.
1 skill от този източник на тази страница
Equity research orchestrator. Routes stock research requests to the right module and enforces analyst discipline: PM seven questions, variant perception, data traceability, rating with target price, and disclaimers. TRIGGER when: the user asks to research/analyze a stock, write an initiation or earn
4 skills от този източник на тази страница
Use this skill after esg-deep-research (and optionally esg-url-scraper), when saved report files for a company need to be grouped by which specific controversy they describe — e.g. "cluster the sources for Anthropic", "group these reports by controversy before assessing". Reads assessment/<company-slug>/reports/*.md frontmatter (including each file's `controversies` labels), groups files that share a label into one cluster, and writes assessment/<company-slug>/reports/clusters.json — which also
Use this skill when asked to research, find, or gather ESG controversy sources for a company from the internet — e.g. "research ESG controversies for {company}", "find sources on {company}'s ESG issues online", "pull articles about {company} controversies". Searches the web for Environmental/Social/Governance controversy coverage, fetches the pages, and saves each source as a Markdown file under assessment/<company>/reports/. Does not classify or score the sources — that is esg-controversy-asses
Use this skill before esg-deep-research, when the user's request needs to be turned into a precise research scope — e.g. "research ESG controversies for Anthropic in 2026", "look at OpenAI and Google since June 2025", "find recent Tesla governance issues". Parses the request to determine, with precision, which company or companies are meant and what date range (start/end) the research should cover, then scaffolds assessment/<company-slug>/{reports,sources,history,scored-text}/ and writes context
Use this skill when asked to scrape, re-fetch, or independently verify ESG research sources for a company that esg-deep-research already ran — e.g. "scrape the sources for {company}", "re-scrape anthropic's sources", "independently fetch {company}'s research URLs". Reads assessment/<company-slug>/reports/manifest.json for the full list of URLs esg-deep-research found, re-fetches every one of them using free no-key scrapers (requests+trafilatura, then Jina Reader as fallback), saves successes int
2 skills от този източник на тази страница
Stand-in skill for the demo-clap example. A real linter ships the skill that explains how to act on its diagnostics, which is the thing an agent cannot work out from the output alone.
Stand-in skill for the demo-lint example. A real linter ships the skill that explains how to act on its diagnostics, which is the thing an agent cannot work out from the output alone.
1 skill от този източник на тази страница
Improve factual reliability while creating or reviewing a course or supplied content. Use when the user asks to fact-check, verify accuracy, reduce hallucinations, make a reliable course, or mentions 事实性错误、知识性错误、专业知识准确性、可靠性. During creation, checks the completed pages before delivery; on existing content, returns a short evidence-backed report and lets the user choose what to fix. Not for grammar, style, or layout. Combine with deep-research when current evidence is the course's main subject.
Пазя съществуващия каталог и добавям намерени SKILL.md файлове след прочит на метаданните. Наличният индекс остава използваем, когато проверката е частична.
Платформата и нивото описват записаното в източника. Те не са гаранция за съвместимост. Български текст се показва само когато в каталога има отделен превод; иначе виждате оригинала с означен език.
Общите звезди показват натрупана популярност. За „Набират интерес“ се сравнява измереният темп при достатъчен интервал. ≥ означава непълна извадка. Под един час не изчислявам дневен темп; методът и изминалият период се показват до записа.
Проверете файла, лиценза и нужния достъп преди употреба. Нито каталогът, нито четенето на SKILL.md изпълняват външни инструкции.
Темата Agent Skills в GitHub