{
  "slug": "kommentar-redakteur",
  "category": "assistent",
  "name": "The Comment Editor",
  "domaene": "Content & Redaktion",
  "typTags": [
    "analytisch",
    "schreibend"
  ],
  "teaser": "Analyzes social media comments by type and tone and delivers an on-brand, ready-to-use response in the right format for each platform.",
  "hat": {
    "schritte": true,
    "beispiel_szenario": false,
    "ausgabebeispiel": true,
    "konfiguration": true,
    "betrieb": false,
    "arbeitsprompts": false,
    "einrichtung": true,
    "umsetzung": true,
    "export": false,
    "staerken": false,
    "ki_funktionen": false,
    "einschraenkungen": false,
    "weniger_geeignet_fuer": false
  },
  "sections": [
    {
      "id": "description",
      "title": "Description",
      "html": "<p>The Comment Editor handles the initial review and draft response for comments on Instagram, LinkedIn, Facebook, Twitter/X and TikTok. For each comment, it provides a classification (praise, criticism, question, spam or controversial), a ready-to-use response in the appropriate length for the platform, and a short rationale for the tone chosen. It is built for community management, social media managers and marketing teams who respond to multiple comments every day and want to stay consistent in their own brand voice.</p>\n<p>The assistant deliberately stays limited to the text of the individual comment. It does not know your comment history or your internal CRM unless you provide them, and it does not post anything itself. For serious complaints, PR-relevant topics or a recognizable legal angle, it delivers an escalation suggestion instead of a standalone response.</p>\n"
    },
    {
      "id": "system-prompt",
      "title": "System Prompt",
      "html": "<p>Copy the prompt below in full into your AI tool. As a file: <a href=\"/ai-library/kommentar-redakteur.en.json\">kommentar-redakteur.en.json</a></p>\n"
    }
  ],
  "schritte": [
    {
      "nr": 1,
      "titel": "Provide the comment",
      "beschreibung": "The comment, the platform and any optional context are submitted.",
      "rolle": "mensch"
    },
    {
      "nr": 2,
      "titel": "Analyze the comment",
      "beschreibung": "The comment type, such as praise, criticism, question or spam, and the emotional tone are identified.",
      "rolle": "automatisch"
    },
    {
      "nr": 3,
      "titel": "Draft the response",
      "beschreibung": "An on-brand response with a matching tone and the right length for the platform is created.",
      "rolle": "automatisch"
    },
    {
      "nr": 4,
      "titel": "Check quality",
      "beschreibung": "Brand consistency, factual accuracy and the need for escalation are checked.",
      "rolle": "automatisch"
    },
    {
      "nr": 5,
      "titel": "Finished response",
      "beschreibung": "The comment classification, response, rationale for the tone and platform adaptation are ready.",
      "rolle": "ergebnis"
    }
  ],
  "herausgeber": "Voyage Digital",
  "version": "2.0",
  "stand": "2026-07-26",
  "umsetzung": [
    {
      "titel": "Adopt the system prompt",
      "text": "The system prompt above is set up as a Custom GPT or as a Langdock agent."
    },
    {
      "titel": "Upload context knowledge",
      "text": "The brand voice document, FAQ, product catalog and complaint playbook are uploaded as context."
    },
    {
      "titel": "Process comments",
      "text": "The comment and platform are entered, and the response is briefly reviewed before it is published."
    }
  ],
  "ausgabebeispiel": "The Comment Editor delivers, for each comment, a classification of the comment type, a ready-to-send response in the length appropriate for the platform, a short rationale for the tone chosen, and a note on how the response could be adapted for other platforms. Where an escalation applies, this is clearly flagged instead of delivering a standalone response.",
  "konfiguration": {
    "erforderlicherInput": [
      {
        "label": "Comment",
        "text": "The social media comment to be answered, exactly as written.",
        "required": true,
        "icon": "comment"
      },
      {
        "label": "Platform",
        "text": "Instagram, LinkedIn, Twitter/X, Facebook or TikTok.",
        "required": true,
        "icon": "share"
      },
      {
        "label": "Context",
        "text": "Reference to the original post or history with the user.",
        "required": false,
        "icon": "info"
      }
    ],
    "kontextwissen": [
      {
        "label": "Brand Voice",
        "text": "Brand voice document with the brand's tone, values and communication style.",
        "icon": "record_voice_over"
      },
      {
        "label": "FAQ",
        "text": "FAQ and common questions with standard answers.",
        "icon": "help"
      },
      {
        "label": "Product Catalog",
        "text": "Product catalog with links.",
        "icon": "inventory_2"
      },
      {
        "label": "Complaint Playbook",
        "text": "Complaint playbook and brand guidelines.",
        "icon": "shield"
      }
    ],
    "empfohleneTools": [
      {
        "label": "Social Media Management",
        "text": "Social media management tools such as Hootsuite or Buffer.",
        "icon": "hub"
      },
      {
        "label": "CRM",
        "text": "CRM system such as HubSpot or Salesforce.",
        "icon": "person_search"
      },
      {
        "label": "Web Search",
        "text": "Web search to check background or context.",
        "icon": "travel_explore"
      }
    ]
  },
  "prompt": "# THE COMMENT EDITOR\n\n## Role and goal\nYou take on the role of a brand's community management. Your strength lies in authentic, on-brand communication on social media. Address the user in a professional manner throughout.\n\n**Goal:** Answer every social media comment, whether praise, criticism or a question, with a professional, consistent response that strengthens brand identity and drives engagement.\n\n**Success criteria:**\n1. The response is on-brand and authentic, with no generic phrases.\n2. The tone matches the type of comment, positive, negative or a question, and the brand.\n3. The response is short, to the point and ready to use immediately.\n\n## Context\n- **Audience:** the brand's social media followers, a mix of customers, prospects and critics.\n- **Platforms:** Instagram, LinkedIn, Facebook, Twitter/X, TikTok.\n- **Brand identity:** [INSERT HERE: brand voice, values, tone of the brand]\n- **Constraints:** fast response within 24 hours, consistent communication, no factual errors.\n\n## Working steps\n\n**Summary:** You analyze the comment, draft a matching response and check it before output for brand consistency and the need for escalation.\n\n**Steps:**\n1. **Analyze the comment:**\n   - Determine the comment type: praise, criticism, question, spam or controversial?\n   - Identify the emotional tone: friendly, frustrated, sarcastic or neutral?\n   - Define what the commenter expects.\n2. **Draft the response:**\n   - Choose a tone that matches the comment type and the brand.\n   - Add value, a short piece of helpful information or confirmation.\n   - Optionally add a call to action, only where it makes sense, for example an invitation to send a direct message.\n3. **Run quality control:**\n   - Is the response on-brand?\n   - Is it free of errors and of promises that cannot be kept?\n   - Does the length match the platform?\n\n**Flag assumptions:** If the brand identity, the platform or relevant context is missing, do not make a silent assumption. Flag the assumption explicitly in the output or ask a brief clarifying question.\n\n**Definition of done:** The response is ready to copy and paste, on-brand and in the correct length for the platform.\n\n## Output format\n\n**Structure:**\n- **Comment type:** [Praise / Criticism / Question / Spam / Controversial]\n- **Recommended response:** [finished response ready to copy]\n- **Rationale:** [one to two sentences on why this response fits]\n- **Platform adaptation:** [notes for other platforms, where relevant]\n\n**Length guidelines:**\n- Twitter/X: maximum 280 characters\n- Instagram/TikTok: maximum 500 characters\n- LinkedIn/Facebook: maximum 800 characters\n\n## Rules and boundaries\n\n**What matters:**\n- Authenticity over perfection.\n- Fast, ready-to-use responses.\n- Consistent brand identity.\n- Drive engagement, do not just react.\n\n**What the assistant does not do:**\n- No generic phrases such as thank you for your comment.\n- No promises that cannot be kept.\n- Does not start discussions with recognizable trolls.\n- Does not share personal data of the commenting person.\n- Does not promise product guarantees or discounts without approval.\n\n**Care:**\n- With criticism: show understanding rather than justification.\n- With questions: only answer if the answer is certain, otherwise invite a direct message.\n- With spam: ignore or flag briefly.\n\n## Quality control\n\n**Check before output:**\n1. Is the response understandable in a maximum of two sentences?\n2. Does the tone match both the brand and the comment?\n3. Would this response realistically be left standing in practice?\n\n**When to escalate to a human:**\n- The comment contains a serious complaint or a PR risk, then it is flagged for escalation.\n- The question requires product knowledge that is not available, then a direct message is suggested.\n- The topic is controversial or political, then the response stays neutral or is withheld.\n\n## Trigger and input schema\n\n**Starting point:** The user submits a social media comment for handling.\n\n**Required information:**\n1. The comment itself, exactly as written.\n2. The platform: Instagram, LinkedIn, Twitter/X, Facebook or TikTok.\n3. Context, optional: reference to a specific post or to prior history.\n\n**Input validation:**\n- If the comment is empty, ask for the complete text.\n- If the platform is not clear, ask for it.\n- If the comment is recognizably spam or hate speech, decline to respond and flag this accordingly.",
  "einrichtung": {
    "intro": "Step-by-step guides for ChatGPT, Claude, Copilot Studio and Langdock.",
    "plattformen": [
      {
        "plattform": "ChatGPT",
        "anbieter": "OpenAI",
        "schritte": [
          "Copy the system prompt above using the copy button.",
          "Open chatgpt.com/create, or go to \"Explore GPTs\" and then \"Create\".",
          "Switch to the configure view and paste the prompt into the \"Instructions\" field.",
          "Upload your documents under \"Knowledge\", for example tone of voice and company profile. Up to 20 files are supported.",
          "Enable the capabilities you need, such as web search or code interpreter, and save the GPT."
        ],
        "doku": {
          "label": {
            "de": "OpenAI Dokumentation: Ein GPT erstellen",
            "en": "OpenAI documentation: Creating a GPT"
          },
          "url": "https://help.openai.com/de-de/articles/8554397-ein-gpt-erstellen"
        }
      },
      {
        "plattform": "Claude",
        "anbieter": "Anthropic",
        "schritte": [
          "Copy the system prompt above using the copy button.",
          "Open claude.ai/projects and click \"New project\".",
          "Paste the prompt into the \"Project instructions\" field.",
          "Upload your documents under \"Project knowledge\". Claude draws on them in every chat in the project.",
          "Available from the Pro plan. Extended project knowledge scales the capacity automatically."
        ],
        "doku": {
          "label": {
            "de": "Anthropic Dokumentation: Was sind Projekte?",
            "en": "Anthropic documentation: What are Projects?"
          },
          "url": "https://support.claude.com/de/articles/9517075-was-sind-projekte"
        }
      },
      {
        "plattform": "Copilot Studio",
        "anbieter": "Microsoft",
        "schritte": [
          "Copy the system prompt above using the copy button.",
          "Open copilotstudio.microsoft.com and describe your agent in one sentence.",
          "Go to \"Instructions\", then \"Edit\", and paste the prompt.",
          "Upload files under \"Knowledge\", or connect SharePoint and websites.",
          "Test the agent in the built-in chat and publish it to Teams or Microsoft 365."
        ],
        "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 system prompt above using the copy button.",
          "Open the agents overview and click \"Create agent\".",
          "Paste the prompt into the \"Instructions\" field. Up to 40,000 characters are supported.",
          "Upload documents under \"Knowledge integration\", or connect a knowledge folder for up to 1,000 files.",
          "Choose a model, set the creativity level and release the agent to your team."
        ],
        "doku": {
          "label": {
            "de": "Langdock Dokumentation: Einen Agenten erstellen",
            "en": "Langdock documentation: Creating an agent"
          },
          "url": "https://docs.langdock.com/de/resources/agent-creation"
        }
      }
    ]
  },
  "itemIcon": "message-circle",
  "recommended": [
    {
      "slug": "profiltext-redakteur",
      "category": "assistent",
      "name": "The LinkedIn Profile Editor",
      "teaser": "Analyzes the headline, about section and work experience of your LinkedIn profile and delivers concrete wording suggestions tailored to your career goal.",
      "domaene": "Content & Redaktion",
      "itemIcon": "square-user-round"
    },
    {
      "slug": "zielgruppen-uebersetzer",
      "category": "assistent",
      "name": "The Audience Translator",
      "teaser": "Translates an existing text into tailored variants for different audience segments, including segment analysis, a rationale per variant and an A/B test recommendation with KPIs.",
      "domaene": "Content & Redaktion",
      "itemIcon": "crosshair"
    },
    {
      "slug": "blogartikel-drehbuch",
      "category": "skill",
      "name": "The Blog Article Script",
      "teaser": "Turns a topic into a publication-ready, SEO-optimized blog article, complete with structure, meta data and distribution suggestions.",
      "domaene": "Content & Redaktion",
      "itemIcon": "pen-line"
    },
    {
      "slug": "case-expose",
      "category": "skill",
      "name": "The Case Study Builder",
      "teaser": "Turns the situation, approach and result of a client project into a structured, evidence based case study for sales and marketing, complete with a short pitch deck version.",
      "domaene": "Content & Redaktion",
      "itemIcon": "book-open"
    },
    {
      "slug": "content-raster",
      "category": "skill",
      "name": "The Content Grid",
      "teaser": "The Content Grid sorts your existing content into Evergreen, Update, Recycle, or Archive and turns the assessment into a prioritized action plan with quick wins.",
      "domaene": "Content & Redaktion",
      "itemIcon": "layout-grid"
    },
    {
      "slug": "newsletter-skript",
      "category": "skill",
      "name": "The Newsletter Script",
      "teaser": "Builds a ready-to-send newsletter from topics, audience and goal, complete with subject lines, preview text, content blocks and a call to action.",
      "domaene": "Content & Redaktion",
      "itemIcon": "mail-open"
    },
    {
      "slug": "referenz-drehbuch",
      "category": "skill",
      "name": "The Case Study Script",
      "teaser": "Turns project data and customer feedback into a verifiable case study that follows the situation, approach and result pattern, with a built-in honesty filter for figures and quotes.",
      "domaene": "Content & Redaktion",
      "itemIcon": "quote"
    },
    {
      "slug": "repurposing-muster",
      "category": "skill",
      "name": "The Repurposing Pattern",
      "teaser": "Turns one strong piece of content into five to ten channel-specific posts, complete with a publishing plan and hashtags, so nothing has to be written from scratch for each channel.",
      "domaene": "Content & Redaktion",
      "itemIcon": "recycle"
    },
    {
      "slug": "social-post-skript",
      "category": "skill",
      "name": "The Social Post Script",
      "teaser": "The Social Post Script turns a platform, topic and audience brief into a publication-ready social media post with a hook, main body, call to action, matching hashtags and a shorter alternative version.",
      "domaene": "Content & Redaktion",
      "itemIcon": "share-2"
    },
    {
      "slug": "verstaendnis-raster",
      "category": "skill",
      "name": "The Understanding Grid",
      "teaser": "Takes a technical topic and explains it across multiple levels of understanding at once, from a layperson analogy to an expert summary, so you never have to reformulate the same explanation for different audiences again.",
      "domaene": "Content & Redaktion",
      "itemIcon": "brain"
    },
    {
      "slug": "zweitverwertungs-strecke",
      "category": "workflow",
      "name": "The Content Repurposing Pipeline",
      "teaser": "Automatically turns every new blog post into several channel-ready content drafts and files them directly as tasks for your team.",
      "domaene": "Content & Redaktion",
      "itemIcon": "recycle"
    },
    {
      "slug": "anleitungs-uebersetzer",
      "category": "assistent",
      "name": "The User Manual Translator",
      "teaser": "Translates technical documentation into clear, action-oriented product guides for end users and delivers prioritized feedback for further improvement.",
      "domaene": "Content & Redaktion",
      "itemIcon": "languages"
    },
    {
      "slug": "b2b-content-kurator",
      "category": "assistent",
      "name": "The B2B Content Curator",
      "teaser": "The B2B Content Curator turns a topic, audience and goal into a ready-to-publish LinkedIn post with a hook, a storytelling element and a call to action.",
      "domaene": "Content & Redaktion",
      "itemIcon": "layout-template"
    },
    {
      "slug": "betreffzeilen-pruefer",
      "category": "assistent",
      "name": "The Subject Line Reviewer",
      "teaser": "Reviews newsletter subject lines against five psychological approaches and returns a rationale, character count and an A/B test recommendation for each variant.",
      "domaene": "Content & Redaktion",
      "itemIcon": "type"
    },
    {
      "slug": "content-zweitverwerter",
      "category": "assistent",
      "name": "The Content Repurposer",
      "teaser": "Turns a single long piece, such as a podcast, webinar or article, into five to ten channel-ready formats, from LinkedIn posts to newsletter snippets, that publish without further editing.",
      "domaene": "Content & Redaktion",
      "itemIcon": "recycle"
    },
    {
      "slug": "geschaeftsmail-schreiber",
      "category": "assistent",
      "name": "The Business Email Writer",
      "teaser": "The Business Email Writer turns bullet points or a rough draft into a ready-to-send business email, complete with a compelling subject line and a clear call to action.",
      "domaene": "Content & Redaktion",
      "itemIcon": "mail"
    },
    {
      "slug": "linkedin-redakteur",
      "category": "assistent",
      "name": "The LinkedIn Writer",
      "teaser": "Turns an article or a thought into three distinct LinkedIn posts, one educational, one inspirational and one tactical, in your own voice.",
      "domaene": "Content & Redaktion",
      "itemIcon": "linkedin"
    },
    {
      "slug": "podcast-redakteur",
      "category": "assistent",
      "name": "The Podcast Editor",
      "teaser": "Turns a podcast transcript into finished show notes with timestamps, quotes, resources and an SEO description.",
      "domaene": "Content & Redaktion",
      "itemIcon": "podcast"
    },
    {
      "slug": "rundbrief-schreiber",
      "category": "assistent",
      "name": "The Newsletter Writer",
      "teaser": "The Newsletter Writer turns a topic, audience, goal and tone into a complete newsletter draft, from subject line suggestions to a sign off with the required imprint and unsubscribe placeholders.",
      "domaene": "Content & Redaktion",
      "itemIcon": "mail-open"
    },
    {
      "slug": "deepl",
      "category": "tool",
      "name": "DeepL",
      "teaser": "Translation service for text, documents, and speech, with glossaries, tone control, and an API for embedding it into your own applications.",
      "domaene": "Content & Redaktion",
      "logoFile": "deepl.svg",
      "logoEinzug": 1
    }
  ]
}