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
54,220 results · page 424 of 1,808
- View details
hyperpod-cluster-debuggerSkillCloud & infra
Lets your agent diagnose and fix problems with HyperPod clusters, like deployment failures and node issues.
Ready to connect★ 895
- View details
hyperpod-issue-reportSkillCloud & infra
Lets your agent collect diagnostic logs and configs from HyperPod clusters to build troubleshooting reports.
Ready to connect★ 895
- View details
hyperpod-ncclSkillAI & models
Lets your agent diagnose GPU cluster training failures like NCCL timeouts, hangs, and version mismatches on HyperPod.
Ready to connect★ 895
- View details
hyperpod-performance-debuggerSkillFiles & storage
Lets your agent diagnose slow SageMaker HyperPod clusters by checking network and filesystem performance signals.
Ready to connect★ 895
- View details
hyperpod-slurm-debuggerSkillProductivity
Lets your agent diagnose Slurm node and job problems on SageMaker HyperPod clusters.
Ready to connect★ 895
- View details
hyperpod-ssmSkillFiles & storage
Lets your agent run commands and move files on SageMaker HyperPod cluster nodes remotely.
Ready to connect★ 895
- View details
librettoSkillWeb & browsing
Lets your agent automate browser tasks by inspecting live web pages and building repeatable workflows.
Ready to connect★ 895
- View details
libretto-readonlySkillWeb & browsing
Lets your agent inspect the live state of a web page in a Libretto browser session without interacting with it.
Ready to connect★ 895
- View details
model-deploymentSkillCloud & infra
Lets your agent generate code that deploys your fine-tuned models to SageMaker endpoints or Bedrock.
Ready to connect★ 895
- View details
model-evaluationSkillAI & models
Lets your agent write Python code to evaluate and benchmark SageMaker models using LLM judges or custom scorers.
Ready to connect★ 895
- View details
promptingSkillAI & models
Helps your agent write and improve system prompts for LLM apps using a minimal, fix-what-fails approach.
Ready to connect★ 895
- View details
pushSkillDev tools
Lets your agent commit and push code changes and open or update a pull request.
Ready to connect★ 895
- View details
push-everythingSkillDev tools
Lets your agent stage every change in the repo, commit, and push it.
Ready to connect★ 895
- View details
sdk-getting-startedSkillCloud & infra
Checks if your computer is set up correctly for SageMaker AI by verifying the SDK version, AWS region, and execution role.
Ready to connect★ 895
- View details
spec-reviewSkillDev tools
Lets your agent review a spec for bugs, gaps, and requirement issues.
Ready to connect★ 895
- View details
use-case-specificationSkillFiles & storage
Lets your agent write a use case spec file defining the business problem, stakeholders, and success criteria for model customization.
Ready to connect★ 895
- View details
userinterface-wikiSkillFiles & storage
Lets your agent review web UI code against UI/UX best practices and report findings with file and line numbers.
Ready to connect★ 895
- View details
skillforgeSkillProductivity
Lets your agent create, improve, find, and quality-check skills for your other agent tasks.
Ready to connect★ 894
- View details
image-crop-rotateSkillFiles & storage
Lets your agent crop images to the center and rotate them 90 degrees clockwise.
Ready to connect★ 893
- View details
pdf-text-replaceSkillDocs & knowledge
Lets your agent fill in or update text like names, addresses, and dates in fillable PDF form fields.
Ready to connect★ 893
- View details
developing-funboost-brokerSkillCommunication
Helps your agent add a custom message broker to the funboost framework by writing Publisher and Consumer classes.
Ready to connect★ 890
- View details
developing-funboost-mixinSkillMonitoring & ops
Guides your agent in writing Mixin extension classes that add monitoring, circuit breaking, rate limiting, and tracing to funboost consumers and publishers.
Ready to connect★ 890
- View details
developing-funboost-testingSkillAI & models
Lets your agent write and run tests for the funboost framework and verify broker or mixin implementations.
Ready to connect★ 890
- View details
funboost-advanced-retrySkillDev tools
Guides your agent to configure advanced retry policies like exponential backoff, dead letter queues, and circuit breakers in funboost.
Ready to connect★ 890
- View details
funboost-async-programmingSkillDev tools
Helps your agent write and fix async (async/await) code in the funboost task framework.
Ready to connect★ 890
- View details
funboost-broker-selectionSkillDatabases & data
Helps your agent pick and configure a message broker (Redis, RabbitMQ, Kafka, MQTT) for funboost tasks.
Ready to connect★ 890
- View details
funboost-faas-deploySkillCloud & infra
Lets your agent turn funboost background tasks into HTTP API endpoints without writing API code.
Ready to connect★ 890
- View details
funboost-funweb-opsSkillCommunication
Lets your agent manage funboost task queues through its web console, checking status and requeueing failed messages.
Ready to connect★ 890
- View details
funboost-memory-queue-poolSkillDocs & knowledge
Lets your agent use funboost's in-memory queue for local concurrency instead of a thread or coroutine pool, with no middleware.
Ready to connect★ 890
- View details
funboost-multi-consumer-startupSkillProductivity
Lets your agent start multiple funboost function consumers at the same time from a single script.
Ready to connect★ 890
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.
53,788 of the 54,220 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.