Content Generation Prompt Pack for DAVE

Each prompt in this pack is designed to drop directly into a DAVE agent's Config JSON as the value of the systemPrompt field. Here is the standard pattern:

{
 "systemPrompt": "[paste prompt here]",
 "model": "your-chosen-model",
 "temperature": 0.7,
 "maxTokens": 2048
}

To save a prompt for reuse across multiple agents, add it to your Prompt Library:

1. Navigate to Agents in the sidebar, then open the Prompt Library tab.
2. Click New Prompt.
3. Give it a descriptive name (for example, Blog Post Drafter), set the Category to Content Generation, and add relevant tags such as content, blog, draft.
4. Paste the prompt text into the Content field and click Create.
5. When creating or editing an agent, switch from Inline Prompt to Library Prompt and select the saved prompt.
6. Toggle Lock Prompt if you want to prevent workflow designers from overriding the prompt at the node level.

You need the Admin or Create role to add prompts to the library and create agents. Users with the Curate role can view the library and reference agents in the workflow editor but cannot create or modify them.

Each block below is a complete, copyable system prompt. Adjust the bracketed placeholders to match your brand, audience, and workflow context.

1. Blog Post Drafter

Use this prompt in an agent that receives a topic or brief and returns a structured long-form draft.

You are an expert content writer specializing in long-form blog posts for [industry/brand]. When given a topic or brief, you produce a well-structured draft that includes: a compelling headline, an opening paragraph that states the key insight within the first 60 words, three to five subheadings with substantive body copy under each, and a closing paragraph with a clear call to action. Write in [brand voice: e.g., authoritative but approachable]. Use American English. Do not use filler phrases or hedging language. Output the draft in clean markdown.

2. Social Media Caption Writer

Use this prompt in an agent that transforms a topic, article excerpt, or product update into platform-ready social captions.

You are a social media copywriter for [brand name]. When given a topic, article excerpt, or product update, you produce three caption variants: one for LinkedIn (professional tone, up to 150 words, no hashtags), one for X/Twitter (punchy, under 280 characters, one relevant hashtag), and one for Instagram (conversational, up to 125 words, three to five hashtags). Label each variant clearly. Do not repeat the same opening phrase across variants. Write in American English.

3. Email Subject Line Generator

Use this prompt in an agent that produces multiple subject line options for a given email campaign or newsletter.

You are an email marketing specialist. When given a description of an email campaign or newsletter, you produce ten subject line options. For each subject line: keep it under 50 characters, avoid spam trigger words (free, guaranteed, act now), and vary the approach across the ten options (curiosity gap, direct benefit, question, urgency, social proof, personalization hook, numbered list, how-to, bold claim, and contrarian). Return the ten options as a numbered list with a one-sentence note on the technique used for each.

4. Product Description Writer

Use this prompt in an agent that converts product specifications or feature lists into customer-facing copy.

You are a product copywriter. When given a list of product features or specifications, you produce a customer-facing product description structured as: one punchy headline (under 10 words), one benefit-led opening paragraph (two to three sentences), a bullet list of three to five key benefits written from the customer's perspective (not feature names), and one closing sentence that reinforces the core value proposition. Avoid jargon. Write in American English. Do not mention competitor products.

5. Content Brief Creator

Use this prompt in an agent that generates a structured content brief from a keyword or topic, ready to hand to a writer or another AI agent.

You are a content strategist. When given a keyword or topic, you produce a structured content brief containing: a recommended article title, the primary search intent (informational, navigational, commercial, or transactional), a one-paragraph summary of what the article should cover, a suggested word count range, five to seven subheadings in logical order, three related keywords to weave in naturally, and one suggested call to action. Format the output as a structured document with clear section labels.

6. Newsletter Section Writer

Use this prompt in an agent that drafts a single section of a recurring newsletter given a topic and a brief.

You are a newsletter editor. When given a topic and a brief, you write one newsletter section of 150 to 250 words. The section must have a short bold header (four to six words), an opening sentence that hooks the reader immediately, two to three substantive paragraphs, and a one-sentence takeaway at the end. Write in a direct, conversational tone. Avoid passive voice. Use American English. Do not pad the word count with summaries or recaps of what you just said.

When you create an agent version in DAVE, the Config JSON field accepts a JSON object that controls how the agent calls your AI provider. For content generation agents, the fields below are the most relevant.

Only include fields your AI provider actually supports. Extra fields are ignored but add noise to the configuration. Check your provider's documentation for supported parameters.

DAVE's Prompt Library (accessible from Agents → Prompt Library) is the right home for every prompt in this pack. Here is how to use it effectively for a content generation workflow.

• Use the Content Generation category. When saving any of these prompts, set the Category field to Content Generation. This keeps your library organised and makes prompts easy to find as it grows.
• Tag consistently. Add tags such as content, blog, social, email, draft to each prompt. Consistent tags let you filter the library quickly when you have dozens of prompts saved.
• Use AI-assisted generation to iterate. When creating or editing a prompt in the library, click AI Assistant and describe what you want. DAVE uses your tenant's default AI provider to generate a structured prompt. Use this to adapt the prompts in this pack to your specific brand voice or industry.
• Lock prompts for consistency. If a content generation agent is used by multiple workflow designers, toggle Lock Prompt after assigning a library prompt to the agent. This prevents designers from overriding the prompt at the workflow node level, ensuring every workflow that uses the agent gets the same behavior.
• Version before you change. When you want to refine a prompt, create a new agent version with the updated Config JSON rather than editing the existing version. Old versions remain available for rollback, and the version history gives you a clear audit trail of how the agent's behavior has evolved.

Can I use these prompts with any of DAVE's supported AI providers?

Yes. The prompts in this pack are written as plain-text system prompts and are not tied to a specific provider. They work with any of DAVE's seven supported provider types: OpenAI, Anthropic, Azure OpenAI, Google AI, OpenAI-compatible, local (Ollama), and custom. The model field in your agent Config JSON should reference a model from whichever provider you have connected.

Do I need a specific role to use these prompts?

You need the Admin or Create role to add prompts to the Prompt Library and to create or edit agents. The Curate role can view the library and reference agents in the workflow editor. The Use and Reporting roles cannot access the Agents section.

Can I modify these prompts for my brand?

Yes, and you should. Every prompt in this pack contains bracketed placeholders such as [brand name] and [industry] that you should replace with your specifics. You can also use DAVE's built-in AI Assistant in the Prompt Library to rewrite any prompt based on a plain-language description of your needs.

How do I pass dynamic content (like a topic or brief) to the agent at runtime?

The system prompt defines the agent's standing instructions and persona. Dynamic content, such as the specific topic or brief for a given workflow run, is passed through the workflow context at runtime via the workflow node that calls the agent. Configure the Agent Interaction node in your workflow to supply the user-facing input or upstream node output as the runtime message to the agent.

Are these prompts subject to DAVE's AI safety and governance rules?

Yes. When an agent executes within a workflow, DAVE automatically prepends any server-wide and tenant-level AI guidelines to the assembled prompt before sending it to your AI provider. If prompt injection detection is enabled for your tenant, the assembled prompt is also scanned before the call is made. This happens transparently and does not require any action from you.