Research & Evidence
Skills
Every public agent skill on file, with what its code can do before you install it. 28,248 skills from 907 public repositories, read Nov 8, 2026. The Monitor links to each skill and never hosts its files.
28,248
skills on file
907
public repositories
4,718
ship scripts
57%
of those touch the network, keys, commands or files
By category
- Other5,128
- Design and media3,994
- Finance and legal3,501
- Code and review3,186
- Data and analysis3,098
- Writing and docs2,113
- Security1,870
- Infrastructure and DevOps1,762
- Productivity1,681
- Marketing and sales1,141
- Research774
What their scripts can do
network1,481
keys or environment1,459
runs commands1,352
deletes files629
Directory
Find a skill
Showing the 48 most copied of 28,248 skills.
karpathy-guidelinesinfrasity-labs/dev-gtm-claude-skillsBehavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring…Writing and docsMITInstructions only47 copiesauth-guidemakenotion/notion-cookbookGuide to setting up third-party authentication for a Notion Worker. Covers brokered credentials for…SecurityMITInstructions only35 copiessyncmakenotion/notion-cookbookScaffold a new sync capability with guided setup — asks about data source, mode, pagination, and cursor…Data and analysisMITInstructions only35 copiessync-debugmakenotion/notion-cookbookDiagnose a failing or misbehaving sync — fetch run logs, identify errors, cross-reference with code, and…Code and reviewMITInstructions only35 copiessync-guidemakenotion/notion-cookbookComprehensive guide to building Notion Workers syncs — covers the two-sync architecture (backfill+delta),…OtherMITnetworkkeys or environment35 copiessync-validatemakenotion/notion-cookbookReview a sync capability for common bugs — cursor advancement, pagination termination, state persistence,…Code and reviewMITInstructions only35 copiescrit-clitomasz-tomczyk/critUse when an agent needs to author or reply to crit inline comments programmatically (including multi-agent…Writing and docsMITInstructions only16 copiescrittomasz-tomczyk/critReview code changes, a plan, a live page (running dev server), or a local HTML file with Crit inline…Code and reviewMITInstructions only12 copiesformat-citationspjt222/agent-almanacFormat citations across academic styles (APA 7, Chicago, Vancouver, IEEE) using CSL processors and R…ResearchMITInstructions only11 copiescareer-opscareer-ops-hq/career-opsOpen-source AI job search agent -- evaluate offers, generate CVs, scan portals, track applications. Use…Marketing and salesMITInstructions only10 copiescrit-storytomasz-tomczyk/critAuthor a crit story and continue the interactive review loop only when the user explicitly invokes…Code and reviewMITInstructions only10 copiesplanning-with-filesOthmanAdi/planning-with-filesPersistent file-based planning for multi-step AI-agent work. Keeps task_plan.md, findings.md, and…ProductivityMITnetworkkeys or environmentruns commandsdeletes files10 copiesadapticpjt222/agent-almanacMaster skill composing the 5-step synoptic cycle for panoramic synthesis across multiple domains.…OtherMITInstructions only10 copiesanalyze-kernel-bottleneckpjt222/agent-almanacSystematically identify whether a GPU kernel is compute-bound, memory-bound, or latency-bound using…Data and analysisMITInstructions only10 copiesanalyze-tensegrity-systempjt222/agent-almanacAnalyze a tensegrity system by identifying compression struts and tension cables, classifying type (class…OtherMITInstructions only10 copiesaudit-discovery-symlinkspjt222/agent-almanacAudit and repair Claude Code discovery symlinks for skills, agents, and teams. Compares registries against…Code and reviewMITInstructions only10 copiesbuild-cli-pluginpjt222/agent-almanacBuild a plugin or adapter for a CLI tool using the abstract base class pattern. Covers defining the…Finance and legalMITInstructions only10 copieschoose-loop-wakeup-intervalpjt222/agent-almanacSelect a `delaySeconds` value when scheduling a loop wakeup via the `ScheduleWakeup` tool or the `/loop`…Writing and docsMITInstructions only10 copiesclean-codebasepjt222/agent-almanacRemove dead code, unused imports, fix lint warnings, and normalize formatting across a codebase without…Data and analysisMITInstructions only10 copiescreate-3d-scenepjt222/agent-almanacSet up a Blender scene programmatically via Python (bpy) with objects, materials, lighting, camera, and…Data and analysisMITInstructions only10 copiescreate-spatial-visualizationpjt222/agent-almanacCreate interactive maps, elevation profiles, and spatial visualizations from GPX tracks, waypoints, or…Data and analysisMITInstructions only10 copiescross-review-projectpjt222/agent-almanacConduct a structured cross-project code review between two Claude Code instances via the cross-review-mcp…Code and reviewMITInstructions only10 copiesdeploy-edge-ai-modelpjt222/agent-almanacDeploy machine learning models to edge devices using Google AI Edge Gallery, TensorFlow Lite, ONNX Runtime,…Infrastructure and DevOpsMITInstructions only10 copiesdesign-cli-outputpjt222/agent-almanacDesign terminal output for a CLI tool with chalk colors, Unicode glyphs, multiple verbosity levels (human,…Design and mediaMITInstructions only10 copiesdesign-electromagnetic-devicepjt222/agent-almanacDesign practical electromagnetic devices including electromagnets, DC and brushless motors, generators, and…Design and mediaMITInstructions only10 copiesdu-dumpjt222/agent-almanacSeparate expensive observation from cheap decision-making in autonomous agent loops using a two-clock…Data and analysisMITInstructions only10 copiesevaluate-agent-frameworkpjt222/agent-almanacAssess an open-source agent framework for investment readiness by evaluating community health, supersession…Finance and legalMITInstructions only10 copiesevaluate-levitation-mechanismpjt222/agent-almanacEvaluate and compare levitation mechanisms for a given application through a structured trade study. Covers…ResearchMITInstructions only10 copiesevolve-skill-from-tracespjt222/agent-almanacEvolve SKILL.md files from agent execution traces using a three-stage pipeline: trajectory collection from…Data and analysisMITInstructions only10 copiesexpand-awarenesspjt222/agent-almanacStructured procedure for expanding attention from focused single-domain mode to panoramic multi-domain…OtherMITInstructions only10 copiesexpress-insightpjt222/agent-almanacCommunicate an integrated insight in a way that is accessible, actionable, and preserves the multi-domain…OtherMITInstructions only10 copiesfile-trademarkpjt222/agent-almanacTrademark filing procedures covering EUIPO (EU), USPTO (US), and WIPO Madrid Protocol (international).…Infrastructure and DevOpsMITInstructions only10 copiesheadless-web-scrapingpjt222/agent-almanacExtract data from web pages using the scrapling Python library — select the appropriate fetcher tier (HTTP,…Data and analysisMITInstructions only10 copiesinstall-almanac-contentpjt222/agent-almanacInstall skills, agents, and teams from agent-almanac into any supported agentic framework using the CLI.…OtherMITInstructions only10 copiesmanage-bibliographypjt222/agent-almanacCreate, merge, and deduplicate BibTeX bibliography files using R packages (RefManageR, bibtex). Parse .bib…ResearchMITInstructions only10 copiesornament-style-modernpjt222/agent-almanacDesign ornamental patterns using modern and speculative aesthetics with colorblind-accessible color scales.…Design and mediaMITInstructions only10 copiesvalidate-referencespjt222/agent-almanacCheck BibTeX entries for completeness, DOI resolution, and broken links. Verify required fields per entry…Design and mediaMITInstructions only10 copiesversion-ml-datapjt222/agent-almanacVersion machine learning datasets using DVC (Data Version Control) with remote storage backends, build…Data and analysisMITInstructions only10 copieswrite-helm-chartpjt222/agent-almanacCreate production-ready Helm charts for Kubernetes application deployment with templating, values…Infrastructure and DevOpsMITInstructions only10 copiessubagent-driven-developmentdavila7/claude-code-templatesUse when executing implementation plans with independent tasks in the current sessionProductivityMITInstructions only9 copiesweb-design-guidelinesdavila7/claude-code-templatesReview UI code for Web Interface Guidelines compliance. Use when asked to "review my UI", "check…Finance and legalMITInstructions only9 copieswriting-plansdavila7/claude-code-templatesUse when you have a spec or requirements for a multi-step task, before touching codeWriting and docsMITInstructions only9 copiesassess-holistic-healthpjt222/agent-almanacConduct temperament-based health assessment from Hildegard von Bingen's Causae et Curae. Evaluates the four…OtherMITInstructions only9 copiesassess-trail-conditionspjt222/agent-almanacEvaluate current trail conditions including weather, snow line, river crossings, exposure, and trail…Writing and docsMITInstructions only9 copiesbuild-consensuspjt222/agent-almanacAchieve distributed agreement without central authority using bee democracy, threshold voting, and quorum…Design and mediaMITInstructions only9 copiesbuild-feature-storepjt222/agent-almanacBuild a feature store using Feast for centralized feature management, configure offline and online stores…Data and analysisMITInstructions only9 copiescatalog-collectionpjt222/agent-almanacCatalog and classify materials using standard library systems. Covers descriptive cataloging, subject…ProductivityMITInstructions only9 copiescenterpjt222/agent-almanacAI dynamic reasoning balance — maintaining grounded reasoning under cognitive pressure, smooth…ProductivityMITInstructions only9 copies
Page 1 of 589
Method
How this list is read
The list starts from Anthropic's own skills repository and the public lists of skills that the community keeps, and reads every repository they link to. Each SKILL.md is read as text, with the scripts in its folder. Nothing is run. A skill that appears in several repositories is listed once, with its copies counted. Licences come from the licence file in the skill's folder or its repository. What the code touches is read from the text of its scripts, so it shows what the code can do, not what it does on any one run.
The Monitor never stores or serves a skill's files: every page links to the author's repository, and the name and description are the author's own words. An author can correct or remove a page through Right of reply.