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
54,680 results · page 581 of 1,823
- View details
engineering-conventionsSkillDev tools
Gives your agent the coding rules and invariants to follow when working on the opencode-swarm codebase.
Ready to connect★ 475
- View details
ensure-all-tests-pass-after-code-changesSkillDev tools
Lets your agent run the test suite and verify all tests pass after code changes.
Ready to connect★ 475
- View details
enterprise-searchSkillSearch
Lets your agent search your company's knowledge across Slack, Confluence, Jira, Notion, GitHub and more with one query.
Ready to connect★ 475
- View details
fork-pr-operationsSkillDev tools
Guides your agent through handling pull requests between forked repositories, including workflow approvals and force-pushes.
Ready to connect★ 475
- View details
gate-attributionSkillDocs & knowledge
Lets your agent assign reviewer and test verdicts from dispatch results to the correct plan tasks.
Ready to connect★ 475
- View details
git-revert-safetySkillDev tools
Guides your agent through safely reverting git commits and undoing merges without breaking related files.
Ready to connect★ 475
- View details
graphifySkillFiles & storage
Lets your agent turn codebases, docs, or media into a knowledge graph to answer questions about structure and relationships.
Ready to connect★ 475
- View details
grep-for-mock-module-across-the-entire-test-file-and-confirm-eveSkillFiles & storage
Lets your agent search test files for 'mock.module' usages and verify each one follows the conversion checklist.
Ready to connect★ 475
- View details
guardrail-patternsSkillDev tools
Lets your agent follow the repo's rules when changing the destructive-command guardrail code in opencode-swarm.
Ready to connect★ 475
- View details
include-test-files-that-assert-on-behavior-being-changed-in-declSkillFiles & storage
Tells your agent to include test files that cover the behavior being changed within the declared scope.
Ready to connect★ 475
- View details
isolate-the-specific-failure-mode-before-proposing-a-fixSkillDev tools
Guides your agent to pinpoint the exact cause of a bug before suggesting any fix.
Ready to connect★ 475
- View details
issue-ingestSkillDev tools
Lets your agent take a GitHub issue, analyze it, and turn it into a clear spec for fixing.
Ready to connect★ 475
- View details
issue-tracerSkillDev tools
Lets your agent investigate, root-cause, fix, and close bug reports, regressions, and failing tests.
Ready to connect★ 475
- View details
linkedin-engagementSkillAI & models
Lets your agent find a LinkedIn post on a topic, draft a comment for your approval, and post it.
Ready to connect★ 475
- View details
merge-queue-readinessSkillDev tools
Lets your agent simulate CI on a PR before adding it to a GitHub merge queue to avoid failed merges.
Ready to connect★ 475
- View details
mock-moduleSkillDev tools
Lets your agent create mock modules for testing code without real dependencies.
Ready to connect★ 475
- View details
mock-to-internals-migrationSkillDev tools
Lets your agent migrate Bun tests from mock.module to injectable internals seams to fix mock leakage.
Ready to connect★ 475
- View details
opt-in-tool-registrationSkillDev tools
Makes your agent ask you before enabling new tools during a session.
Ready to connect★ 475
- View details
orchestrating-subagentsSkillAI & models
Guides your agent on delegating tasks to helper subagents, choosing the right model and effort for each role.
Ready to connect★ 475
- View details
parallel-work-checkSkillAI & models
Lets your agent check if others are already working on a branch before it starts, avoiding duplicate or conflicting changes.
Ready to connect★ 475
- View details
phase-completeSkillDev tools
Lets your agent mark a development phase as complete when working with the opencode-swarm workflow.
Ready to connect★ 475
- View details
phase-wrapSkillAI & models
Lets your coding agent mark and manage phase boundaries when running multi-agent coding workflows.
Ready to connect★ 475
- View details
pr-readinessSkillDev tools
Redirects your agent to the swarm-pr-feedback skill, replacing this older pull-request feedback tool.
Ready to connect★ 475
- View details
pre-phase-briefingSkillProductivity
Lets your agent run a structured briefing at the start of a task, reviewing evidence and checking readiness.
Ready to connect★ 475
- View details
propagation-testSkillDev tools
Lets your agent test event or data propagation across a swarm of code agents.
Ready to connect★ 475
- View details
qa-sweepSkillDev tools
Lets your agent run an independent QA review of code changes before publishing.
Ready to connect★ 475
- View details
research-firstSkillDocs & knowledge
Makes your agent research current external facts and docs before planning or implementing a task.
Ready to connect★ 475
- View details
restore-internals-seams-in-finally-blocks-after-each-testSkillDev tools
Lets your agent automatically restore internal code seams in finally blocks after each test runs.
Ready to connect★ 475
- View details
run-affected-test-file-in-isolation-and-in-co-run-with-its-test-SkillFiles & storage
Lets your agent rerun the test file touched by code changes, alone and together with its related tests.
Ready to connect★ 475
- View details
run-affected-test-suite-end-to-end-after-multi-file-signature-chSkillFiles & storage
Lets your agent run the affected test suite end-to-end after signature changes across multiple files.
Ready to connect★ 475
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.
54,249 of the 54,680 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.