What are agent skills? Reuse the way your team works
Skills package the way your team performs a task. Instead of rebuilding an audit, brief, or research method in every prompt, attach a reusable skill to the agents that should follow it.
skills
Reusable Runbook
runbooks
3
avg steps
12
SEO audit
Check rankings, search intent, competitors, and next-page bets.
Weekly report
Pull metrics, explain movement, draft the leadership summary.
Action brief
Turn findings into owners, experiments, and approval gates.
Last run: SEO audit · 2d ago
The concept
A plain-language definition before the product implementation.
A skill is a reusable procedure, not a tool credential
A skill contains instructions and supporting files that teach an agent how to approach a recurring task. It can describe the sequence, quality bar, templates, and references. Connections provide access to external tools; skills explain how the work should be done.
How AI Agent uses it
Create once, attach deliberately
AI Agent lets you create or import skills, inspect their source and files, and attach them to one or more agents. A skill remains inactive until attached, which keeps the agent's usable procedures explicit.
Reusable runbooks
Capture the steps and quality bar for work your team repeats.
Supporting files
Keep templates and references alongside the main skill instructions.
Visible attachments
See which agents use a skill and add or remove attachments centrally.
From idea to a working system.
- 01
Author or import
Create a skill manually or bring one in from a supported source.
- 02
Attach
Choose the agents that should be able to use the procedure.
- 03
Reuse
Select the skill in chat or let an equipped agent apply it during work.
Package the method when the team repeats the work
Best for
Reusable procedures, templates, references, and quality bars that several agents should apply consistently.
Choose another pattern when
Live data access or credentials. A connection grants tool access; a skill explains how an equipped agent should work.
Included in AI Agent
The product capabilities behind the idea.
- Markdown skill instructions
- Supporting files
- Source provenance
- Many-to-many agent attachments
Frequently asked questions
Keep exploring
Related product guides
Agents
An agent is the reasoning layer: it understands a goal, decides which approved tools to use, and works toward an outcome. AI Agent gives that reasoning a clear operating contract instead of relying on one giant prompt.
Workflows
A workflow turns a business process into an explicit graph. The model can reason inside an agent step, while the surrounding sequence keeps triggers, tools, branches, approvals, and outputs predictable.
Connections
Connections give agents access to the tools where work and data already live. Access stays separate from the agent's instructions and knowledge so credentials, allowed tools, and review rules remain controlled.
Build the smallest useful version first.
Start with the agent and one real task. Add workflow control, context, and approvals when the work shows you where they matter.