# Primary source

Concept · [Building with AI](https://ecosystem.samcarlton.com/ai)

Also called: primary sources, raw source, original source

The original material itself, such as the code, a full [transcript](https://ecosystem.samcarlton.com/speech-to-text) or the raw data. It is complete and trustworthy but often large, so loading it costs [tokens](https://ecosystem.samcarlton.com/token) and [attention](https://ecosystem.samcarlton.com/attention).

**For example:** To find out how the signup form really behaves, the [agent](https://ecosystem.samcarlton.com/ai-agent) reads the code, the primary source, instead of an old [summary](https://ecosystem.samcarlton.com/secondary-source).

## Connects to (5)

- [Secondary source](https://ecosystem.samcarlton.com/secondary-source): A description of the original material rather than the material itself, such as a summary, a README or a compaction note.
- [Context window](https://ecosystem.samcarlton.com/context-window): How much text an LLM can take into account at one time.
- [Token](https://ecosystem.samcarlton.com/token): In AI, a small chunk of text, roughly a word or a piece of a word, that an LLM reads and writes.
- [Attention budget](https://ecosystem.samcarlton.com/attention-budget): A way of picturing attention as a fixed amount each token can spread across everything else in the input.
- [Contextual knowledge](https://ecosystem.samcarlton.com/contextual-knowledge): Facts the model can read directly in front of it right now, such as documents you pasted, files the agent opened or search results.

Map view: https://ecosystem.samcarlton.com/primary-source