There are several ways to add prompts to your Lindy:
Lindy-level Settings: These are overarching instructions that Lindy keeps in mind throughout the entire automation process. This prompt is added as a prefix to every LLM call made by this Lindy, across all steps, providing crucial context about the overall task. Example: "You are an AI assistant for a marketing agency. Always maintain a professional tone and focus on marketing-related tasks." Action Parameters: These are specific instructions Lindy uses to complete a particular field within a given action. They allow for precise control over individual steps in your workflow. Example: "Extract the customer's name and email address from the following text, formatting them as 'Name: [name], Email: [email]'." AI Agents: AI Agents are ideal when the next steps in a process are uncertain or highly variable. They operate using broad guidelines and a set of defined skills that can be invoked as needed. The agent decides which skills to use and when, based on the guidelines provided. Example guideline: "You are a customer service agent. Your goal is to resolve customer inquiries efficiently while maintaining a friendly tone." 1. Structure Your Prompts Clearly Be Direct Skip the pleasantries. Lindies understand and execute direct commands best.
Example:
You must take the value in column F and cross-reference it with the email address captured in the body of the webhook.
Provide a Persona and Define the Content Give Lindy a clear persona and define her role for each task. This helps set the tone, perspective, and level of expertise for her responses.
Tailor your instructions to this persona, which will guide Lindy in understanding the context and required expertise.
When appropriate, especially for customer-facing communications, advise Lindy to use natural, conversational language. This approach ensures that Lindy's responses are not only accurate but also aligned with the specific requirements of each task and audience.
Example:
You are an expert at B2B sales. Craft a compelling 2-sentence follow-up email that includes a hook and a strong call-to-action (CTA). All communication with the customer should be conversational and friendly.
Action-Focused Use active language. Prefer "do" over "don't" to clearly convey desired actions.
Example:
Always add this email signature at the end of every message: [insert signature here]
Step-by-Step Break down complex tasks into simple, sequential steps, guiding Lindy through each part of the process.
Example:
1. Analyze the webpage text.
2. Identify all open job openings.
3. Qualify the job openings, selecting only those that are remote positions.
2. Specify What You Need Include References If your task is based on a template or previous example, instruct Lindy to use the same language or structure. Leverage Lindy's ability to maintain context throughout the workflow by referencing data from prior steps.
Example:
Use the job openings from your web search (Step 2) to update the Google Document. Include only the Front End Job Postings in this update.
Keywords and Phrases Incorporate essential words to emphasize important aspects and priorities. Use phrases like:
"Your task is..." "You MUST..." "It is crucial that..." "You will be penalized if..." 3. Customize Content and Style Consistency is Key Instruct Lindy to maintain the flow and style of writing from previous steps in the workflow or examples you have added to the prompt. This ensures a cohesive output across the entire process.
Edit for Clarity Every word matters! Proofread your prompts for grammar, punctuation, and vocabulary to ensure they are clear and consistent. Ambiguity can lead to unexpected results.
4. Use Delimiters Effectively Delimiters in prompts serve several important functions:
Organize information Emphasize specific sections Reduce ambiguity Improve consistency Enhance readability Common types of delimiters include:
Triple quotes: """example text""" Triple backticks: example text Triple dashes: ---example text--- XML-like tags: <example>example text</example> Hashtags: ###example text### 5. Clear Formatting Separate instructions with line breaks for clarity. This visual organization helps Lindy process information more effectively.
6. Enhance Specificity and Information Provide complete information for tasks that require detailed responses or actions. Context is crucial for accurate results.
Example: If Lindy is responding to customer support tickets, provide context about your customer base and the types of inquiries they typically make.
7. Final Tips Test and Iterate After setting your instructions, test them thoroughly and refine them based on the outcomes. Continuous improvement is key to optimal performance.
Keep Learning As you use Lindy, you'll discover new ways to improve your instructions. Regularly update your guidelines to reflect these insights and best practices.
Support and Community If you're ever unsure or need additional guidance, don't hesitate to reach out. Email support@lindy.ai for assistance.
