Skills.

Give your AI a better way to work.

A skill is a set of written instructions that teaches an AI how to do one job the way it should be done: review a pull request, plan a migration, write the release notes.

Install one here and it travels with your account into Claude, Claude Code, Cursor and every other client you sign in with.

Filter by category

55,914 results · page 282 of 1,864

  • ltx2SkillMedia

    Lets your agent generate video clips from text or images, including b-roll and animated backgrounds.

    Ready to connect★ 2k

    github.com/digitalsamba/claude-code-video-toolkit2k stars

    View details
  • moviepySkillFiles & storage

    Lets your agent edit and compose videos in Python, adding captions and overlays to AI-generated clips.

    Ready to connect★ 2k

    github.com/digitalsamba/claude-code-video-toolkit2k stars

    View details
  • playwright-recordingSkillWeb & browsing

    Lets your agent record browser interactions as video using the playwright claude skill for demos and walkthroughs.

    Ready to connect★ 2k

    github.com/digitalsamba/claude-code-video-toolkit2k stars

    View details
  • qwen-editSkillMedia

    Guides your agent in writing effective prompts for editing photos with Qwen-Image-Edit, like changing clothes or poses.

    Ready to connect★ 2k

    github.com/digitalsamba/claude-code-video-toolkit2k stars

    View details
  • runpodSkillCloud & infra

    Lets your agent set up and manage cloud GPU jobs on RunPod, including deploying images and checking costs.

    Ready to connect★ 2k

    github.com/digitalsamba/claude-code-video-toolkit2k stars

    View details
  • evSkillProductivity

    Lets your agent list pull requests, issues, and other project info for the Evennia project.

    Ready to connect★ 2k

    github.com/evennia/evennia2k stars

    View details
  • add-featureSkillDev tools

    Lets your agent add claude skill features to Syncpack, like new CLI commands and validation logic.

    Ready to connect★ 2k

    github.com/jamiemason/syncpack2k stars

    View details
  • document-codeSkillWeb & browsing

    Lets your agent update the Syncpack documentation website when you make user-facing code changes.

    Ready to connect★ 2k

    github.com/jamiemason/syncpack2k stars

    View details
  • fix-bugSkillDev tools

    Debugs and fixes Syncpack bugs using hypothesis-driven testing when tests fail or behavior is unexpected.

    Ready to connect★ 2k

    github.com/jamiemason/syncpack2k stars

    View details
  • front-loadingSkillDocs & knowledge

    Lets your agent write clearer docs and messages by putting the key answer first, then details after.

    Ready to connect★ 2k

    github.com/jamiemason/syncpack2k stars

    View details
  • optimize-benchmarksSkillDev tools

    Lets your agent run benchmarks, find performance bottlenecks, apply optimisations, and verify improvements in a loop.

    Ready to connect★ 2k

    github.com/jamiemason/syncpack2k stars

    View details
  • search-codeSkillSearch

    Lets your agent search the Syncpack codebase for symbols and code patterns using grep and ast-grep.

    Ready to connect★ 2k

    github.com/jamiemason/syncpack2k stars

    View details
  • signal-over-noiseSkillDocs & knowledge

    Lets your agent rewrite text to cut filler and hedging so docs and messages are clear and concise.

    Ready to connect★ 2k

    github.com/jamiemason/syncpack2k stars

    View details
  • testing-gapsSkillFiles & storage

    Lets your agent analyze which parts of a source file lack test coverage and identify missing test scenarios.

    Ready to connect★ 2k

    github.com/jamiemason/syncpack2k stars

    View details
  • write-codeSkillDev tools

    Guides your agent to write claude skill Rust code that follows Syncpack's style rules for imports, naming, and quality.

    Ready to connect★ 2k

    github.com/jamiemason/syncpack2k stars

    View details
  • backward-compatibility-policySkillFiles & storage

    Guides your agent to refactor code by dropping backward-compatibility shims, following the KHI project's policy.

    Ready to connect★ 2k

    github.com/googlecloudplatform/khi2k stars

    View details
  • googlecloud-apiSkillDev tools

    Teaches your agent the required patterns for calling Google Cloud APIs correctly in KHI code.

    Ready to connect★ 2k

    github.com/googlecloudplatform/khi2k stars

    View details
  • jj-workflowSkillDev tools

    Lets your agent manage code changes with Jujutsu, including workspaces, commits, PR feedback, and conflicts.

    Ready to connect★ 2k

    github.com/googlecloudplatform/khi2k stars

    View details
  • khi-parserSkillMonitoring & ops

    Guides your agent to add new log type parsers or modify existing ones in the KHI log inspection tool.

    Ready to connect★ 2k

    github.com/googlecloudplatform/khi2k stars

    View details
  • log-timeline-mapperSkillMonitoring & ops

    Guides your agent in writing and testing log ingestion and timeline mapping code for KHI inspections.

    Ready to connect★ 2k

    github.com/googlecloudplatform/khi2k stars

    View details
  • streaming-apiSkillDev tools

    Guides your agent in designing server-side streaming RPCs with polling fallbacks following KHI conventions.

    Ready to connect★ 2k

    github.com/googlecloudplatform/khi2k stars

    View details
  • task-record-replaySkillProductivity

    Lets your agent create and run microbenchmarks and pprof profiling for KHI DAG tasks.

    Ready to connect★ 2k

    github.com/googlecloudplatform/khi2k stars

    View details
  • nemotron-3-ultra-text2sql-loraSkillDev tools

    Lets your agent run an end-to-end tutorial that fine-tunes a large language model for text-to-SQL on a SLURM cluster.

    Ready to connect★ 2k

    github.com/nvidia-nemo/nemotron2k stars

    View details
  • nemotron-add-modelSkillAI & models

    Lets your agent onboard a new AI model into the Nemotron skills library so other skills can route tasks to it.

    Ready to connect★ 2k

    github.com/nvidia-nemo/nemotron2k stars

    View details
  • nemotron-add-patternSkillDev tools

    Lets your agent add a reusable ML decision pattern file that other planning steps can follow.

    Ready to connect★ 2k

    github.com/nvidia-nemo/nemotron2k stars

    View details
  • nemotron-add-stepSkillDev tools

    Lets your agent scaffold a new pipeline step with manifest, configs, and README in the nemotron steps directory.

    Ready to connect★ 2k

    github.com/nvidia-nemo/nemotron2k stars

    View details
  • nemotron-customizeSkillAI & models

    Lets your agent plan and run Nemotron model customization pipelines like fine-tuning, alignment, and optimization.

    Ready to connect★ 2k

    github.com/nvidia-nemo/nemotron2k stars

    View details
  • nemotron-customizer-airgapSkillCloud & infra

    Lets your agent prepare and validate offline airgap image bundles for Nemotron Customizer deployments.

    Ready to connect★ 2k

    github.com/nvidia-nemo/nemotron2k stars

    View details
  • nemotron-nano3SkillCloud & infra

    Lets your agent answer factual questions about the Nemotron Nano 3 model, like its architecture, training, and deployment.

    Ready to connect★ 2k

    github.com/nvidia-nemo/nemotron2k stars

    View details
  • nemotron-policy-generatorSkillDocs & knowledge

    Lets your agent draft custom content-safety policies and category taxonomies for Nemotron guardrails.

    Ready to connect★ 2k

    github.com/nvidia-nemo/nemotron2k stars

    View details

What is a skill?

A skill is plain text, usually a SKILL.md file and the scripts it refers to, written for an AI rather than for a person. It carries the steps, the house rules and the examples a good answer needs, so you stop pasting the same briefing into every new chat.

55,483 of the 55,914 skills listed here can be served through ahel today, and they come from public repositories. Each one has its own page with the instructions themselves on it, so you can read what a skill will tell your AI to do before you install it.

Install one and every AI you use gets it

Installing a skill adds it to your gateway and turns it on in the same step. Claude Code surfaces it as a slash command; any client can read the full instructions with the skill_read tool.

Nothing is copied into a project folder. The instructions are served from your account, so the same skill is there in every AI you connect, and turning it off removes it from all of them at once.

See how to connect your AI