Skip to content
UseAgent

Reproducible Procedures

Your procedures.
Always available.

Give recurring tasks a consistent starting point with reusable instructions and team procedures.

Procedure match

Review this PR for security issues

  1. 1PR-review-checklistgithub.com/myorg/playbooks94%
  2. 2Security-audit-templategithub.com/myorg/playbooks87%
  3. 3Code-standardsgithub.com/myorg/playbooks71%
Illustrative example

Reusable team playbooks

Your best practices as reusable skills. GitHub-sourced. Auto-synced. Semantically ranked.

GitHub import + auto-sync

Scan a repository for SKILL.md files. Import selected paths as versioned skills pinned to a commit. Optional hourly resync keeps the catalog current without manual work.

Semantic turn-start ranking

Every unpinned turn, the agent sees skills ranked by relevance to the prompt. Right procedures surface without fixed top-N limits. You control the source: GitHub is truth.

Playbooks use the same substrate

Playbooks are skills with kind=playbook. Same versioning, same import, same prefill ranking. No duplicate infrastructure.

SEE IT WORK

One agent. Real tasks.

UseAgentExample workflow

Our widget service returns 500 on DELETE /widgets/:id when the widget has 3 or more linked components. Fix it.

  • Clone and search

    Cloned repo, found DELETE handler in handlers/widgets.ts. Searched error logs. Every 500 has 'permission check failed' after component count check.

  • Reproduce locally

    Set up test DB with 3 linked components, sent DELETE request, confirmed 500. Traced the stack: permission middleware runs after cascade check, causing the 500.

  • Fix the order

    Moved permission check before cascade. Added test case: DELETE with 3+ components now returns 204. Old test suite still green.

  • Open PR

    Created PR #487: 'Fix widget delete with linked components'. Added test case to prevent regression.

Questions, answered.

Do agents always use the top-ranked skill?

No. Ranking shows what's relevant, but agents choose the exact skill to activate via the skill_activate tool. Agents can skip a high-ranked skill if it doesn't fit the task.

Can I write skills outside GitHub?

You can create them manually in UseAgent, but GitHub import is the designed path. Manual skills don't auto-sync.

What's the difference between a skill and a playbook?

They're the same substrate with different presentation. Playbooks are typically end-to-end workflows, skills are smaller procedures. Use whichever label fits your content.

Start with one task.
See what you can hand off.

Explore a completed run, or join the early-access list to hear about getting your team started.