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
54,221 results · page 320 of 1,808
- View details
golem-scala-base-imageSkillMedia
Lets your agent understand the Golem Scala SDK WIT worlds and rebuild its QuickJS guest runtime WASMs.
Ready to connect★ 2k
- View details
golem-scala-code-generationSkillDev tools
Lets your agent develop Scala source generation code for the Golem build plugins.
Ready to connect★ 2k
- View details
golem-scala-developmentSkillDev tools
Lets your agent compile, test, and publish the Golem Scala SDK and its sbt/Mill build plugins.
Ready to connect★ 2k
- View details
golem-scala-integration-testsSkillAI & models
Lets your agent run and debug integration tests for a Scala SDK.
Ready to connect★ 2k
- View details
golem-schedule-agent-moonbitSkillProductivity
Lets your agent schedule a delayed or timed future call to a MoonBit Golem agent from the command line.
Ready to connect★ 2k
- View details
golem-schedule-agent-rustSkillProductivity
Lets your agent schedule one of its own tasks to run at a future time.
Ready to connect★ 2k
- View details
golem-schedule-future-call-moonbitSkillProductivity
Lets your agent schedule another agent call at a future time, like setting a timer or delay.
Ready to connect★ 2k
- View details
golem-schedule-future-call-rustSkillProductivity
Lets your agent schedule its own tasks to run later, after a delay or at a set time.
Ready to connect★ 2k
- View details
golem-skill-harnessSkillAI & models
Lets your agent write, test, and debug skill scenarios for Golem using a TypeScript test harness.
Ready to connect★ 2k
- View details
golem-stateless-agent-moonbitSkillAI & models
Lets your agent create stateless Golem agents in MoonBit that start fresh on each request with no shared state.
Ready to connect★ 2k
- View details
golem-stateless-agent-rustSkillAI & models
Helps your agent build stateless agents in Rust that start fresh on each call with no shared state.
Ready to connect★ 2k
- View details
golem-test-crash-recoverySkillMonitoring & ops
Lets your agent simulate a crash so you can test that it recovers and resumes correctly.
Ready to connect★ 2k
- View details
golem-trigger-agent-moonbitSkillAI & models
Lets your agent start a Golem agent task in the background without waiting for it to finish.
Ready to connect★ 2k
- View details
golem-trigger-agent-rustSkillAI & models
Lets your agent start a fire-and-forget run of a Rust Golem agent method without waiting for results.
Ready to connect★ 2k
- View details
golem-troubleshoot-buildSkillFiles & storage
Helps your agent diagnose Golem build failures and fix golem.yaml manifest configuration problems.
Ready to connect★ 2k
- View details
golem-undo-agent-stateSkillAI & models
Lets your agent undo recent operations and roll back its state to recover from errors.
Ready to connect★ 2k
- View details
golem-update-running-agentsSkillCloud & infra
Lets your agent update running applications to new component versions.
Ready to connect★ 2k
- View details
golem-view-agent-filesSkillFiles & storage
Lets your agent list and download files from a running agent's virtual filesystem.
Ready to connect★ 2k
- View details
golem-view-agent-logsSkillMonitoring & ops
Lets your agent view and stream logs and output from running agents.
Ready to connect★ 2k
- View details
golem-wait-for-external-input-moonbitSkillAI & models
Lets your agent pause and wait for external input or events using Golem promises.
Ready to connect★ 2k
- View details
golem-wait-for-external-input-rustSkillAI & models
Lets your agent pause a Rust workflow until an external signal or data arrives.
Ready to connect★ 2k
- View details
investigating-benchmark-performanceSkillMonitoring & ops
Lets your agent run benchmarks with tracing enabled and analyze Jaeger traces to find slow spots.
Ready to connect★ 2k
- View details
investigating-executor-performanceSkillMonitoring & ops
Lets your agent diagnose slow executor tests by enabling OTLP tracing and analyzing trace spans in Jaeger.
Ready to connect★ 2k
- View details
managing-docs-versionsSkillDocs & knowledge
Lets your agent cut new documentation versions, promote next to a release, and manage versioned docs content.
Ready to connect★ 2k
- View details
migrate-ts-decorator-sdkSkillAI & models
Helps your agent rewrite old TypeScript Golem agent code using the removed decorator API to the current defineAgent API.
Ready to connect★ 2k
- View details
modifying-builtin-pluginsSkillDev tools
Lets your agent build and modify built-in plugins, including their descriptors, versions, and provisioning.
Ready to connect★ 2k
- View details
modifying-cli-manifest-schemaSkillDev tools
Lets your agent modify an application manifest JSON schema and update CLI schema references.
Ready to connect★ 2k
- View details
modifying-cli-output-schemaSkillDev tools
Lets your agent modify how the Golem CLI formats and structures its command output data.
Ready to connect★ 2k
- View details
modifying-http-endpointsSkillDev tools
Lets your agent add or change HTTP REST API endpoints in Golem services.
Ready to connect★ 2k
- View details
modifying-service-configsSkillDev tools
Lets your agent change service configuration structs, fields, and default values.
Ready to connect★ 2k
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.
53,789 of the 54,221 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.