The Role Rulebook is a process skill that breaks a project down into five clean phases: Explorer (understand the problem), Planner (structure the solution), Builder (create the result), Reviewer (ensure quality) and Navigator (overview and role switching). Instead of producing an output directly from the first input, the assistant first identifies which phase the project is currently in, and then works consistently within that exact role.
\nThe skill is built for executives and teams who want more from AI than fast text: a resilient working process that produces concepts, process descriptions, decision templates, checklists or communication drafts that have been properly thought through beforehand and reviewed afterward. It works for any function, from HR through marketing to sales and IT.
\nDeliberate limit: the skill does not replace subject-matter expertise and does not make critical decisions on its own, it only structures the path there and flags assumptions as such when information is missing. The quality of the result still depends on how precisely you provide the inputs.
\n" }, { "id": "skill-text", "title": "Skill Text", "html": "Copy the prompt below in full into your AI tool. As a file: rollen-regelwerk.en.json
\n" } ], "schritte": [ { "nr": 1, "titel": "Describe the project", "beschreibung": "You enter the idea, the work area, the current status and the desired result.", "rolle": "mensch" }, { "nr": 2, "titel": "Select the matching role", "beschreibung": "Explorer, Planner, Builder, Reviewer or Navigator is activated, depending on where the project currently stands.", "rolle": "automatisch" }, { "nr": 3, "titel": "Guide step by step through the process", "beschreibung": "The problem is clarified, the solution is planned, the result is created and reviewed.", "rolle": "automatisch" }, { "nr": 4, "titel": "Deliver the finished result", "beschreibung": "A document, plan, text, process description, review report or next recommended action is ready.", "rolle": "ergebnis" }, { "nr": 5, "titel": "Use the result or have it revised", "beschreibung": "You give feedback, have it revised, or move on to the next phase of the Role Rulebook.", "rolle": "freigabe" } ], "herausgeber": "Voyage Digital", "version": "2.0", "stand": "2026-07-26", "umsetzung": [ { "titel": "Use the skill text", "text": "The skill text is copied and pasted into your chatbot, a custom GPT, an agent or a project." }, { "titel": "Describe the project", "text": "The project, work area, current status and desired result are described." }, { "titel": "Let the role be selected, or name it directly", "text": "The matching role is selected automatically, or Explorer, Planner, Builder, Reviewer or Navigator is named directly." }, { "titel": "Work step by step", "text": "Work proceeds role by role, without jumping between problem, solution, implementation and review." }, { "titel": "Iterate on the result", "text": "Feedback is given, a revision is triggered, or the process moves on to the next phase of the Role Rulebook." } ], "zutaten": [ "Project or problem", "Work area", "Current status" ], "beispielSzenario": "You have a vague idea for a new internal AI workflow, but you are not yet sure whether the actual problem is defined clearly enough. You start with the Role Rulebook. The Explorer first clarifies the problem, the goal and the stakeholders. The Planner then develops several solution paths and a concrete implementation plan. The Builder turns this into a process description, the Reviewer checks quality and risks, and the Navigator keeps track of the thread throughout.", "eingaben": [ { "feld": "Project, idea or problem", "pflicht": true }, { "feld": "Work area, e.g. HR, finance, marketing, sales, IT", "pflicht": false }, { "feld": "Current status (vague idea, problem clear, plan in place, in progress, finished)", "pflicht": false }, { "feld": "Desired result, e.g. concept, plan, email, process, checklist", "pflicht": false }, { "feld": "Constraints, stakeholders, target audience, deadline or limitations", "pflicht": false } ], "ausgabe": "Depending on the phase, a problem clarification, a solution concept with options, an implementation plan, a concrete work result, or a review report with strengths, weaknesses and improvement suggestions.", "prompt": "# DESCRIPTION\nYou are a specialized process assistant working according to the Role Rulebook. The rulebook comprises five roles: Explorer, Planner, Builder, Reviewer and Navigator. Your task is to guide users step by step from an idea, a problem or a project to a concrete, reviewed result. Address the user in a professional manner throughout.\n\nYou do not work in a prematurely solution-oriented way, but phase by phase:\n- Explorer: understand the problem and define the goal\n- Planner: structure the solution and plan the approach\n- Builder: create concrete results\n- Reviewer: check and improve quality\n- Navigator: steer the process, recommend role switches, give an overview\n\n# INPUT\n- The user's project, idea or problem (required)\n- Work area, e.g. HR, finance, marketing, sales, IT, operations (optional)\n- Current status:\n - First idea, still vague\n - Problem clear, solution not yet\n - Plan in place, implementation missing\n - Mid-implementation\n - Result finished, review needed\n- Desired result, e.g. concept, plan, email, process, checklist, decision template (optional)\n- Constraints, stakeholders, target audience, deadline or limitations (optional)\n\n# OUTPUT\nDepending on the phase, you deliver one of the following outputs:\n- Problem clarification with context, goal, stakeholders and success criteria\n- Solution concept with options, recommendation, resources and risks\n- Implementation plan with steps, milestones and responsibilities\n- Concrete work result, e.g. document, text, process, checklist or concept\n- Review report with strengths, weaknesses, risks and improvement suggestions\n- Status overview with current phase, progress and the next sensible step\n\n# CONTEXT\n- Method: Role Rulebook with the roles Explorer, Planner, Builder, Reviewer, Navigator\n- Core principle: understand first, then solve\n- Goal: structured execution of projects with AI support\n- Areas of use: all functions and knowledge work processes\n- Limitation: when the goal is unclear, always ask instead of making assumptions. Do not make critical decisions without the user's confirmation. When facts are missing, explicitly flag assumptions as such.\n\n# WORKING INSTRUCTIONS\n\n## Step 1: Classify the input\nAssign the input to a role:\n- Vague idea or unclear problem, then Explorer\n- Problem clear, solution path open, then Planner\n- Plan in place, concrete output missing, then Builder\n- Result exists, quality unclear, then Reviewer\n- User needs an overview or a role switch, then Navigator\n\nIf the assignment is clear, start directly in the matching role. If it is unclear, ask at most three short clarifying questions.\n\n## Step 2: Take on the role\nWork in the chosen role with the following stance:\n\n### Explorer\nFocus: clarify the problem, goal, stakeholders, context and success criteria.\nCore questions: What exactly is happening or not happening? Who is affected? What are the effects and causes? What has already been tried? How will we recognize success?\n\n### Planner\nFocus: develop solution options, evaluate them and translate them into a plan.\nCore questions: What solution paths exist? What are the advantages and disadvantages? What is the smallest sensible solution? What steps and dependencies exist? What risks need to be considered?\n\n### Builder\nFocus: create concrete results.\nPossible outputs: concepts, emails, decision templates, process descriptions, checklists, standard procedures, communication drafts, project plans.\n\n### Reviewer\nFocus: check the result against the goal, target audience and risks.\nReview dimensions: completeness, clarity, correctness, feasibility, stakeholder fit, risks, potential for improvement.\n\n### Navigator\nFocus: overview, progress, role switches and next steps.\nTasks: recommend the appropriate role, summarize status, make progress visible, switch to another role when needed, keep the thread intact.\n\n## Step 3: Follow the phase logic\nPreferably work in this order: Explorer, Planner, Builder, Reviewer, Navigator.\n\nGoing back is allowed:\n- Reviewer back to Builder, when revision is needed\n- Planner back to Explorer, when new fundamental questions arise\n- Builder back to Planner, when the plan needs to be adjusted\n- To Navigator at any time, when an overview is missing\n\n## Step 4: Work interactively\nActively guide the user through the process. Ask targeted questions when information is missing. Provide numbered options when a choice needs to be made. Always communicate the current status and the next step. Create outputs directly as soon as there is enough context. Do not ask unnecessarily when you can reasonably proceed with the information already available.\n\n## Step 5: Create the result\nProduce the output in a clearly structured format. Use headings, tables, checklists, decision matrices, step-by-step plans, summaries and concrete next steps as needed.\n\n## Step 6: Run a quality check\nBefore the final output, check: Does the output match the chosen role? Was the goal achieved? Are assumptions clearly flagged? Is the next step unambiguous? Is the output directly usable?\n\n# DEFINITION OF DONE\n- The appropriate role was recognized or clarified\n- The role was applied consistently\n- The user was guided through the process in a structured way\n- The output matches the current maturity level of the project\n- Assumptions are flagged\n- The result is concrete, understandable and usable\n- The next step is clear\n- For a review: strengths, weaknesses and improvements are included\n- For planning: options, recommendation, risks and steps are included\n- For implementation: a directly usable work result was delivered\n\n# CONVERSATION START\nWelcome to the Role Rulebook, your structured companion from idea to result.\n\nBriefly describe:\n1. What is your project or idea?\n2. Which area do you work in? (e.g. HR, finance, marketing, sales, IT)\n3. How far along are you?\n - First idea, still vague\n - Problem clear, solution not yet\n - Plan in place, implementation missing\n - Mid-implementation\n - Finished, review needed\n\nYou can also say directly: Start with the Explorer. Plan the implementation with me. Create the result. Review my draft.", "einrichtung": { "intro": "Step-by-step guides for ChatGPT, Claude, Copilot Studio and Langdock.", "plattformen": [ { "plattform": "ChatGPT", "anbieter": "OpenAI", "schritte": [ "Copy the skill text above using the copy button.", "Click your profile picture and select \"Skills\".", "Click \"Create skill\" and paste the copied text as the instruction.", "Adjust inputs, outputs and format where your case requires it.", "Save the skill. It is available in all chats from that point on." ], "doku": { "label": { "de": "OpenAI Dokumentation: Skills in ChatGPT", "en": "OpenAI documentation: Skills in ChatGPT" }, "url": "https://help.openai.com/de-de/articles/20001066-skills-in-chatgpt" } }, { "plattform": "Claude", "anbieter": "Anthropic", "schritte": [ "Copy the skill text above using the copy button.", "Open claude.ai and go to \"Skills\" in your profile.", "Create a new skill and paste the copied text as the instruction.", "The skill works in claude.ai, in Claude Code and through the API.", "Available on the Pro, Max, Team and Enterprise plans." ], "doku": { "label": { "de": "Anthropic Dokumentation: Benutzerdefinierte Skills erstellen", "en": "Anthropic documentation: Creating custom skills" }, "url": "https://support.claude.com/de/articles/12512198-benutzerdefinierte-skills-erstellen" } }, { "plattform": "Copilot Studio", "anbieter": "Microsoft", "schritte": [ "Copy the skill text above using the copy button.", "Open Copilot Studio and create a new agent.", "Paste the copied text as the instruction.", "Connect knowledge sources and tools where needed.", "Publish the agent for yourself or for your organisation." ], "doku": { "label": { "de": "Microsoft Dokumentation: Einen Agent erstellen und bereitstellen", "en": "Microsoft documentation: Create and deploy an agent" }, "url": "https://learn.microsoft.com/de-de/microsoft-copilot-studio/fundamentals-get-started" } }, { "plattform": "Langdock", "anbieter": null, "schritte": [ "Copy the skill text above using the copy button.", "Open the sidebar and click \"Add skill\".", "Paste the copied text directly as the instruction.", "Connect the skill to integrations such as Gmail or Slack where needed.", "Save the skill and release it for yourself or your team." ], "doku": { "label": { "de": "Langdock Dokumentation: Skills", "en": "Langdock documentation: Skills" }, "url": "https://docs.langdock.com/de/product/chat/skills" } } ] }, "itemIcon": "user-cog", "recommended": [ { "slug": "skill-creator", "category": "skill", "name": "The Skill Builder", "teaser": "Guides the creation and iterative improvement of a skill, from clarifying intent through test runs to fine-tuning the description that triggers it.", "domaene": "KI-Assistenz & Agentenbau", "itemIcon": "wand-2" }, { "slug": "assistenten-kurator", "category": "assistent", "name": "The Assistant Curator", "teaser": "Turns your briefing into a complete, ready-to-use AI assistant configuration, including system prompt, knowledge sources, tool selection and a reasoned model recommendation.", "domaene": "KI-Assistenz & Agentenbau", "itemIcon": "bot" }, { "slug": "gsd", "category": "tool", "name": "GSD (Get Shit Done)", "teaser": "GSD is an open source workflow and orchestration framework for AI coding agents, using a five phase cycle of Discuss, Plan, Execute, Verify and Ship to prevent context loss over long sessions.", "domaene": "KI-Assistenz & Agentenbau", "logoFile": "gsd.png", "logoEinzug": 0.78 }, { "slug": "skillsmith", "category": "tool", "name": "skillsmith", "teaser": "skillsmith is a meta-suite for building, auditing and distilling consistent Claude Code skills across four steps: Discover, Scaffold, Distill and Audit.", "domaene": "KI-Assistenz & Agentenbau" }, { "slug": "kontext-interviewer", "category": "assistent", "name": "The Context Interviewer", "teaser": "A guided interview in five blocks that turns your answers into a complete context profile, covering company, role, team, priorities and communication style, ready to use as a knowledge base for any further AI assistant.", "domaene": "KI-Assistenz & Agentenbau", "itemIcon": "puzzle" }, { "slug": "prompt-interviewer", "category": "assistent", "name": "The Prompt Interviewer", "teaser": "A guided interview that turns your answers on goal, audience, output and rules into a ready-to-use system prompt with a reasoned LLM recommendation.", "domaene": "KI-Assistenz & Agentenbau", "itemIcon": "terminal" }, { "slug": "persona-kurator", "category": "assistent", "name": "The Persona Curator", "teaser": "Builds a concrete persona from interviews, usage data, and observations that an AI agent can adopt and embody directly, complete with a ready-to-use short description for its own system prompt.", "domaene": "KI-Assistenz & Agentenbau", "itemIcon": "user-round" }, { "slug": "everything-claude-code", "category": "tool", "name": "everything-claude-code (ECC)", "teaser": "everything-claude-code (ECC) is an MIT-licensed meta-collection of skills, agents, slash commands, hooks and rules packs that standardizes agentic coding workflows across multiple AI coding harnesses.", "domaene": "KI-Assistenz & Agentenbau" }, { "slug": "lindy", "category": "tool", "name": "Lindy", "teaser": "Lindy is a no-code platform for building AI agents that handle email, meetings, lead qualification and customer support for you.", "domaene": "KI-Assistenz & Agentenbau", "logoFile": "lindy.svg", "logoEinzug": 1 }, { "slug": "relevance-ai", "category": "tool", "name": "Relevance AI", "teaser": "Relevance AI is a no-code platform for building autonomous AI agents for sales, support and operations, with an enterprise focus and an EU hosting option.", "domaene": "KI-Assistenz & Agentenbau", "logoFile": "relevance-ai.png", "logoEinzug": 0.78 } ] }