Skills
Save a method for one of your agent's jobs so it can follow the same steps and checks on future work.
Keep the method, then use it on new work
A skill saves the steps and checks behind a good result. Your agent can use that method again with new facts, dates, and sources without you repeating the same coaching.
A skill is a reusable method for a particular kind of work. It can tell an agent where to look, what steps to follow, how to check the result, and what finished work should include.
An agent has a broader role and may handle several jobs. A customer-relationship agent might prepare for calls, draft follow-ups, and review renewals. Its call-brief skill explains how to do the first of those jobs; other skills can guide the others.
Skills are different from memories, which hold lasting facts and preferences, and from tools, which let an agent take an action. A skill tells the agent how to use information and tools for the job.
What skills can teach
Skills can cover a process, a way to make decisions, or the shape of a deliverable. A useful skill may combine these.
A process
The steps for preparing a call brief, checking a quote, or turning research into a report.
A way to decide
Which sources to trust, what counts as a risk, and when to ask for missing information.
A deliverable format
What to include in a brief, email, webpage, or report, with examples and checks.
A tool workflow
How to use a connected service or API for a job, including what to do when information is missing.
Your agents arrive already trained on the basics
Hyperagent includes skills for built-in work. Your own skills add the methods and standards specific to your team.
From thread to skill
Skills often begin with work you and an agent have already improved together. Once the result meets your needs, the repeatable steps and checks can become a skill. Customer details, current numbers, and temporary deadlines stay with the job rather than becoming part of the method.
In the thread, ask Hyperagent to propose a skill from the work. Review the draft, save it, and make it available to the agent that needs it. Teach your agent how you work walks through those actions and a fresh-work test.
A call-brief skill might capture which customer records to check, how to identify open issues, and how to structure the brief. The next request supplies the actual customer and meeting date.
Anatomy of a skill
Before a skill can help an agent, it needs a clear label, a clear purpose, and enough instruction for the agent to follow the method.
A Call Brief skill should say when to use it, which sources to check, what the brief must include, and what to flag for review. A specific name helps the agent find it when the job calls for it.
Name and description
What you and the agent use to recognize the skill. Specific names are easier to find and safer to pin.
When to use
Situations where the method should apply. This is the main signal for discovery when the skill is not pinned.
Documentation
Steps, examples, constraints, inputs, and what good output looks like. This is the core of the skill.
Scripts and credentials
Code for repeatable mechanics, plus credential fields for services those scripts need. Values stay out of the conversation. See Skills with APIs and credentials.
How an agent uses skills
Once a skill exists, the agent needs a way to bring it into the thread. That is controlled by Skill Discovery, pin state, and how far the agent is allowed to search.
How skills enter a thread
Pinned always, discovered when needed, or preloaded as a fixed kit.
Pinned skills are visible from the start. The agent sees the skill's name and description at the beginning of the thread, so it can recognize when that method applies. A customer-relationship agent might keep Call Briefs pinned if preparing for meetings is central to its role.
With Skill Discovery on, other skills can be found when needed. The agent starts lean: pinned summaries first, then search when the task points at a method. Full documentation loads only when matched. Context stays available for the conversation.
With Skill Discovery off, attached skills are preloaded. The agent starts with full documentation for its attached skills and does not search broadly. That works well for a small, fixed kit. It uses more context from the first turn.
How agents find the right skill
With discovery on, clear fields matter. A skill named "Call Briefs" with when-to-use text like "Use when preparing for a customer meeting" is easier to match than a generic "Reporting" skill.
If you name the skill in your message, the agent can search for it and load the full instructions when its scope allows that search. You do not need to pin every method to use it.
Search scope
Search scope controls how far the agent can look when a skill is not already visible.
All skills
The agent can discover skills across your knowledge base, even when they are not attached. Use when a well-built method should be available without wiring it to every agent by hand.
Attached only
The agent only finds skills you linked in agent settings. Use for shared or Slack-facing agents that should not pick up methods from unrelated workflows.
In agent settings, the Skill Discovery toggle usually drives these choices together: on means discover with broader search; off means preload the attached set. Full detail on how skills load and how the discovery boundary works lives in Knowledge and learning.
Share and improve a method
A skill can help more than one agent when those agents have access to it and the method fits their jobs. A call-brief skill might help a customer-relationship agent and another agent preparing a leadership update, while each keeps its own role and standards.
Share a skill through a workspace when teammates should maintain or use the same method. Check who owns the skill, which agents can discover it, and whose credentials it uses for connected work. Improving the shared source can help future runs that use it; work already in progress does not automatically reload new instructions. Fork the skill when a team needs an independent variation.