Knowledge · assist
AI Skills: explain recurring tasks once, use them permanently
What AI Skills are, which processes are suitable for them, and how to teach your AI once how your company works.
AI Skills are abilities that you teach an AI once: recurring processes, rules, or task patterns of your company – such as "This is how we write quotes" or "This is how we respond to complaints". Once stored as a Skill, no one has to repeat the explanation: neither you nor your colleagues, and the AI doesn't keep asking for the same basics in every conversation.
The problem: good prompts get lost
Anyone who works with AI regularly knows the pattern: for one task, after some trial and error, a really good instruction emerges – tone, structure, rules, examples. The next time it's gone, buried in an old chat history or in a text file that only one person on the team knows about. The result: the same explanatory work happens again and again, and the quality varies depending on who's asking.
Skills solve exactly that: they turn a fleeting prompt into a permanent, maintained ability.
How a Skill is structured
In webRichtung assist, a Skill has a clear structure:
- a name and a type
- an area it belongs to (e.g. strategy, technology)
- a priority and a status (e.g. Active)
Added to this is the distinction between your own Skills and the Skills of your organization: personal Skills reflect your way of working, organization Skills apply to the whole team. This way the assistant answers a complaint for the new colleague according to the same rules as for the owner.
Which tasks are good Skill candidates
The rule of thumb: good candidates are things you regularly explain the same way. For example:
- Texts with fixed rules: quotes, rejections, payment reminders, reply emails to typical inquiries
- Processes with steps: how a complaint is recorded, how an order is documented
- Formats: how a weekly report, a protocol, or a handover is structured
- House rules: form of address, tone, what may not be promised, when a human has to take over
Unsuitable are one-off special cases – you explain those faster in a conversation than you maintain a Skill.
How to build a good Skill
- Observe: Which explanation are you giving for the third time? That's your candidate.
- Be specific: Not "write good quotes", but structure, mandatory details, tone, and a good example.
- Name the boundaries: What the Skill may not decide – such as discounts or legal commitments.
- Test and refine: Check the results, write deviations back into the Skill.
- Share: Turn proven Skills into an organization ability so the whole team benefits.
Why the effort pays off
A Skill costs a one-time half hour of care – and afterwards saves explanation time with every use. More importantly: it preserves knowledge that otherwise depends on individuals. How your company writes quotes is then no longer only in the head of the most experienced colleague, but a maintained ability in the system – a building block of what distinguishes an AI chat with company knowledge from a generic chatbot. How to create Skills is shown in the documentation at docs.webrichtung.de/assist/.
FAQ
What is an AI Skill?
An ability that you teach your AI: a recurring process, a rule, or a task pattern of your company – defined once, then usable without explaining again.
Which tasks are suitable as a Skill?
Anything you regularly explain the same way: how you write quotes, how you answer complaints, how a weekly report is structured.
Do Skills apply only to me or to the whole team?
Both make sense: there are personal Skills and organization Skills that apply to the whole team – so everyone works according to the same patterns.
How does a Skill differ from a good prompt?
A prompt applies to one conversation, a Skill permanently. You don't have to repeat the explanation – the knowledge is stored and maintained in one place.