Turn chat ideas into ready to use AI prompts and save them in Airtable. Great for teams that build n8n agents and want a clean prompt library with names and categories. A simple chat message starts it.

A chat message triggers the flow. Google Gemini writes a new prompt from your input. A second Gemini step reads that prompt and uses a structured parser to return clean JSON with a name and a category. An auto fixing parser repairs small format issues so the record stays valid. Set nodes map fields, then a create step adds the record to your Airtable Prompt Library with Name, Prompt, and Category. The result is also sent back to the chat.

You need n8n Cloud, a Google Gemini API key, and an Airtable base with the three fields. Expect faster prompt creation and better tagging, since every entry follows the same structure. Many teams can use it, like internal AI builders, operations, and support content owners. It cuts manual work and builds a searchable library in minutes. Setup takes a short time and pays off as your prompt list grows.

• Chat message trigger captures prompt ideas instantly
• Gemini 2.0 Flash Lite generates a fresh prompt from your text
• Structured Output Parser enforces name and category in JSON
• Auto fixing parser repairs common formatting errors
• Set nodes map fields for Name, Prompt, and Category
• Airtable create action writes a new record to your selected base and table
• Final response is returned to the chat for quick review

• Reduce manual prompt writing from 30 minutes to 2 minutes
• Improve tag consistency by 90 percent with structured parsing
• Handle 10 times more prompt ideas without extra staff
• Connect Google Gemini and Airtable in one flow
• Keep a single source of truth for prompts and categories

1. Import the template into n8n: Create a new workflow in n8n, click the three dots menu, select 'Import from File', then choose the downloaded JSON file
2. You'll need accounts with Google Gemini and Airtable. See the Tools Required section above for links to create accounts with these services.
3. In the n8n editor, double click the Google Gemini Chat Model nodes. In the Credential to connect with dropdown, click Create new credential and follow the on screen steps. Create or paste your Google Gemini API key from the Google AI Studio API page, name the credential, and save.
4. Open the Airtable node. In the Credential to connect with dropdown, click Create new credential and choose Airtable Personal Access Token. In your Airtable account, create a personal access token with data read and write scopes for the target base, paste it into n8n, name the credential, and save.
5. In Airtable, prepare a base and table with fields named Name, Prompt, and Category. Use single line text fields for all three to match this template.
6. Back in n8n, in the Airtable node, pick your base and table from the dropdowns. Confirm the column mapping so Name, Prompt, and Category point to the right fields. Keep operation as Create.
7. Open the Generate a new prompt node and review the system message. Adjust tone or style if you want different prompt writing rules.
8. Open the Structured Output Parser node and confirm the schema includes name and category. Leave it as is to ensure clean JSON.
9. Check the Set nodes to confirm text, name, category, and prompt mappings are correct. Save the workflow.
10. Open the n8n AI Chat panel and send a sample idea to start the flow. Keep the workflow in test mode to see live executions.
11. Troubleshoot common issues: If no Airtable record appears, verify the token scopes and field names. If parsing fails, confirm the schema and that the Auto fixing parser is connected. Review n8n execution logs for error details.