Additional workstation / team day
An additional intensive day or the setup of further workstations, to extend the setup to a whole team. Extends the enablement in breadth, without changing the scope of the first day.
available now
Service · Wolff Applied AI
Agentic AI: Your AI Agent in a Day
From the basic app to a productive agentic workflow in one intensive day — tools set up safely, real use cases implemented, limits honestly stated.
1 intensive day + preparation/follow-up (~1–2 weeks elapsed)Duration
On requestInvestment
6Deliverables
Why this offer matters now
On one focused intensive day — with structured preparation and follow-up — we guide you from the mere AI app to a productively used agentic workflow. We choose the fitting tool provider-neutrally (Claude Code, Codex, Cursor or Google Antigravity), set it up safely and clarify subscriptions, limits and costs as well as the guardrails on legal, data-protection and compliance matters — what the agent is allowed to do and which approvals are needed. On the intensive day we work hands-on: prompting and work instructions, skills/plugins, project memory (CLAUDE.md, AGENTS.md, memory files) and subagents, and we productively implement one to two real use cases. We honestly state limits and risks throughout. The result is a working, safely configured agentic workflow that you can continue in everyday work — not a tool rollout, but a genuine way of working.
Starting point
Agentic AI tools — Claude Code, Codex, Cursor, Google Antigravity — shift work from "asking a question" to "delegating a task": the agent reads context, plans steps and executes them with approvals. The potential is considerable, but getting started often fails at three hurdles. First, selection and safe setup: which tool fits, which subscription, which configuration is privacy-friendly? Second, security: what is the agent allowed to do, which approvals and limits are needed, how does confidential data stay protected? Third, productive anchoring: without guided first use cases, project memory and clear work instructions, it stays at the experimentation stage. A structured intensive day with preparation and follow-up overcomes these hurdles together — honest in assessing limits and risks.
Your benefit
- A working, safely configured agentic workflow that you see in use on the same day and can continue independently afterwards.
- Provider-neutral selection of the fitting tool (Claude Code, Codex, Cursor, Google Antigravity) together with a clear assessment of subscriptions, limits and costs.
- Safe installation and configuration with clear approvals and limits — what the agent is allowed to do, what not, and how confidential data stays protected.
- One to two real use cases are productively implemented on the intensive day — directly visible benefit instead of an abstract demo.
- Fundamentals that carry: prompting and work instructions, skills/plugins, project memory (CLAUDE.md, AGENTS.md, memory) and subagents.
- An honest assessment of limits, risks and costs — as a robust basis for the next step towards productive operation.
A good fit when …
- Your teams already use AI apps safely and want to take the next step towards agentic workflows.
- You want to select and set up an agentic tool (Claude Code, Codex, Cursor, Google Antigravity) safely and to fit.
- You need clear guardrails on what an agent is allowed to do, which approvals are needed and how data stays protected.
- You want to experience the productive benefit on one concrete, real use case — instead of staying at general demos.
- You want limits, risks and costs honestly assessed before rolling out more broadly.
How we work
4 phases, 4 gates — you decide at every transition.
- 01
P0 · Preparation & Scoping Call
In a preliminary conversation we clarify goals, existing accounts/subscriptions, technical framework conditions and one to two suitable real use cases. We choose the fitting tool provider-neutrally and set the security and approval framework for the intensive day.
Gate — your approvalG0: tool selected, one to two real use cases established, technical and legal framework conditions (approvals, data protection) for the intensive day clarified. - 02
P1 · Setup & Security
We set up the chosen tool safely: installation, privacy-friendly configuration, subscription/limit settings as well as clear approvals and limits (what the agent is allowed to do). Legal, data-protection and compliance guardrails are recorded in an understandable way.
Gate — your approvalG1: tool working and safely configured, approvals/limits documented, data-protection guardrails agreed — before productive work on the real use case begins. - 03
P2 · Intensive Day (Hands-On)
On the intensive day we work together productively: prompting and work instructions, skills/plugins, project memory (CLAUDE.md, AGENTS.md, memory files) and subagents. One to two real use cases are implemented to a usable result; limits and risks are honestly assessed throughout.
Gate — your approvalG2: at least one real use case productively implemented and demonstrated working, core building blocks (work instructions, project memory) applied. - 04
P3 · Follow-Up & Anchoring
We hand over configuration, work instructions/prompts and a short guide, honestly record limits, risks and costs, and give a recommendation for the next steps — from an agent office hour to controlled productive deployment (AI Pilot Support).
Gate — your approvalG3: materials and configuration handed over, limits/risks/costs documented, those responsible trained, next step placed.
What you receive
- Tool selection & setup documentation
A reasoned, provider-neutral tool choice (Claude Code/Codex/Cursor/Google Antigravity) with the installation and configuration status as well as subscription/limit/cost assessment.
- Security & approval guardrails
Clearly documented limits and approvals of the agent as well as data-protection/compliance guardrails — organisational orientation, not legal advice.
- Productively implemented use cases
One to two real use cases, implemented to a usable result on the intensive day and demonstrated working.
- Work instructions, prompts & project memory
Developed work instructions/prompts as well as an established project memory (CLAUDE.md, AGENTS.md, memory files) and, where applicable, subagents as a reusable basis.
- Skills/plugins overview
Assessment and initial setup of fitting skills/plugins for the addressed tasks, with pointers for extension.
- Limits, risk & next-steps recommendation
An honest assessment of limits, risks and costs as well as a reasoned recommendation for anchoring and the further path (office hour, team day, pilot support).
Base package plus optional modules
The base package stands on its own. Optional modules dock on now or later — you decide how far you go.
Agent office hour (4 weeks)
An accompanying office hour over around four weeks after the intensive day catches practical questions and consolidates the new way of working. Useful to book later, once first everyday experience is available.
available later
Controlled productive deployment
The path from the proven workflow to reliable productive operation — with success criteria, gates and operational handover. Mapped as a separate follow-on via the AI Pilot Support (Offer 10).
available later
Good to know & honest scope
Good to know
Prerequisites
- Basic familiarity of participants with AI apps (ideally from the AI Basic Enablement, Offer 13).
- A workstation/device with the necessary rights to install or use the chosen tool, as well as existing or to-be-procured subscriptions/accounts (provided by the client).
- One to two suitable, approved real use cases and the necessary (non-sensitive or approved) materials for them.
Your involvement
- Naming a technical and a business contact person for scoping, approvals and the intensive day.
- Involvement of IT and data protection for installation rights, approved tools and approval limits.
- Providing the real use-case context and timely approvals, so that productive work is possible on the intensive day.
Credit: A preceding AI Basic Enablement (Offer 13) can be credited proportionally when commissioned directly.
Honest scope
- –Subscription/licence and usage costs of the tools (e.g. model- or API-dependent fees) — these are borne by the client.
- –The broad rollout and productive ongoing operation of agentic workflows — the AI Pilot Support (Offer 10) is intended for that.
- –A legally binding data-protection/EU AI Act assessment: the security and compliance guardrails are organisational orientation, not legal advice.
- –Individual software development beyond the guided use-case implementation (can be mapped separately as an effort item).
Note: not legal advice. Content with a legal dimension (e.g. EU AI Act, GDPR, ISO/IEC 42001) is professional orientation, not legal counsel.
Key facts
You receive a clear, individual proposal before any engagement. Under § 19 UStG no VAT is charged; any ancillary costs (e.g. model/hosting/API fees, travel) are not included.
Next step
Ready to get concrete?
Let us clarify in a free intro call whether this offer fits your situation — honestly and without sales pressure.As a follow-up we then recommend AI Pilot Guidance.
Non-binding · online or on site · reply < 24 h