All templates By NappAI team Tweet publishing automation
Post a tweet based on content previously posted on the user's account.
graph TD
%%{init: {'theme': 'mc','layout': 'elk'}}%%
ParseData-a5xpo[<div><img alt="logo" src="/_astro/braces.Djq0PW4_.svg" style="height: 20px !important;width: 20px !important"/></div>Parse Data]
style ParseData-a5xpo stroke:#a170ff
Prompt-2wdmr[<div><img alt="logo" src="/_astro/square-terminal.BMOXc-nZ.svg" style="height: 20px !important;width: 20px !important"/></div>Prompt]
style Prompt-2wdmr stroke:#a170ff
OpenAIModel-djdyd[<div><img alt="logo" src="/_astro/openAI.BhmuxEs3.svg" style="height: 20px !important;width: 20px !important"/></div>OpenAI]
style OpenAIModel-djdyd stroke:#a170ff
XReader-tbiw6[X Reader]
style XReader-tbiw6 stroke:#a170ff
XPoster-ob91z[X Poster]
style XPoster-ob91z stroke:#a170ff
Prompt-2wdmr -.- OpenAIModel-djdyd
linkStyle 0 stroke:#a170ff
ParseData-a5xpo -.- Prompt-2wdmr
linkStyle 1 stroke:#a170ff
XReader-tbiw6 -.- ParseData-a5xpo
linkStyle 2 stroke:#a170ff
OpenAIModel-djdyd -.- XPoster-ob91z
linkStyle 3 stroke:#a170ff
🧩 Overview
The workflow automates the creation and publication of a new tweet based on a user’s historical activity on X. It retrieves recent tweets, formats the data into a simple text representation, generates a fresh tweet using an OpenAI model, and finally posts the generated content back to X. This end‑to‑end process streamlines social‑media content creation and ensures the new tweet remains coherent with the account’s existing voice and topics.
⚙️ Main Features
- Retrieves recent tweets from a specified X user using the X API.
- Parses raw tweet data into a plain‑text summary suitable for prompt consumption.
- Builds a prompt that instructs the LLM to produce an innovative, concise tweet aligned with past themes.
- Generates tweet text with an OpenAI language model under controlled temperature and token limits.
- Publishes the generated tweet to the user’s X timeline via the X API.
🔄 Workflow Steps
| Component Name | Role in the Workflow | Key Inputs | Key Outputs |
|---|---|---|---|
| X Reader | Retrieves tweet data from X | Username, Max Results, Credential ID | Tweets Data |
| Parse Data | Converts the tweet data into a plain‑text format using a specified template | Tweets Data | Parsed Text |
| Prompt | Builds the prompt that will guide the LLM, injecting the parsed tweet history | Parsed Text, Max Characters | Prompt Message |
| OpenAI | Generates the tweet text from the prompt | Prompt Message, Model Settings | Generated Tweet Text |
| X Poster | Publishes the generated tweet to X | Generated Tweet Text, Credential ID | Confirmation / Tweet Data |
🧠 Notes
- The workflow requires valid X API credentials and an OpenAI API key; credentials must be configured in the corresponding components.
- The OpenAI model runs with a fixed temperature of 0.1 to ensure high‑coherence, but users may adjust this to allow more creative output.
- Rate limits on both the X API and the OpenAI service apply; excessive executions may result in temporary suspension of access.
- The Parse Data component expects the X Reader to return data in the default JSON format; custom schemas may require adjustments to the template.
- Generated tweets are limited to 100 characters by default, but this can be altered via the Prompt component’s
max_charactersfield. - Privacy considerations: tweet content is sent to the OpenAI service; ensure compliance with data‑handling policies before deploying.