AI-Powered RSS Workflow
Start by opening AI Agents in Make and creating a new agent. Make allows you to use built-in AI without connecting external providers, which is enough for simple workflows. External models such as OpenAI or Gemini can be added later if needed. Name the agent MMA Journalist, select the Large model, paste in the system prompt from the previous chapter, and save.
After the agent is created, additional settings become available. One important concept is Context. Context allows you to upload reference files that the agent can rely on instead of guessing. This is especially useful for business or support workflows where accuracy matters. For this RSS-to-tweet example, no context is required, so it can be left empty.
Context
Context allows uploading reference material such as text files, PDFs, or documents. The agent can rely on this data instead of guessing.
When context matters
- FAQs;
- Product details;
- Policies;
- Any scenario where accuracy is critical.
Context improves reliability, but it is not required for this RSS-to-tweet workflow.
Passing Inputs to the Agent
Define the message content by mapping fields from the RSS module:
- Title;
- Description;
- Summary;
- Link or URL.
Even if the agent does not browse links, including the URL keeps the input structured.
Thanks for your feedback!
Ask AI
Ask AI
Ask anything or try one of the suggested questions to begin our chat
Awesome!
Completion rate improved to 4.35
AI-Powered RSS Workflow
Swipe to show menu
Start by opening AI Agents in Make and creating a new agent. Make allows you to use built-in AI without connecting external providers, which is enough for simple workflows. External models such as OpenAI or Gemini can be added later if needed. Name the agent MMA Journalist, select the Large model, paste in the system prompt from the previous chapter, and save.
After the agent is created, additional settings become available. One important concept is Context. Context allows you to upload reference files that the agent can rely on instead of guessing. This is especially useful for business or support workflows where accuracy matters. For this RSS-to-tweet example, no context is required, so it can be left empty.
Context
Context allows uploading reference material such as text files, PDFs, or documents. The agent can rely on this data instead of guessing.
When context matters
- FAQs;
- Product details;
- Policies;
- Any scenario where accuracy is critical.
Context improves reliability, but it is not required for this RSS-to-tweet workflow.
Passing Inputs to the Agent
Define the message content by mapping fields from the RSS module:
- Title;
- Description;
- Summary;
- Link or URL.
Even if the agent does not browse links, including the URL keeps the input structured.
Thanks for your feedback!