# Docs (https://docs.neuralisapp.com/package-system/docs)

`docs/*.md` is the category for material the agent should be able to consult
without obeying it as a constraint: API references, background, lookup tables.
Docs are **not** injected into the prompt. They are listed in the package
overview by title and description and read on demand, the same lazy model as a
skill body. This category is also the landing spot for anything
trust-sanitized out of `rules/` (see [Rules](https://docs.neuralisapp.com/package-system/rules.md)).

## How the agent reads a doc [#how-the-agent-reads-a-doc]

- A built-in package's doc is read through the agent-core inspection skill's
  `files/content` call
  (`neuralis request GET @neuralis/agent-core/files/content -f package=<package> -f category=<category> -f id=<id>`),
  which serves only an active file the caller's own catalog lists.
- A project-installed or source package's doc is read with `fs_read` on its URI.

An author can force a single doc into the prompt with frontmatter
`activation.pinned` or `activation.target: "prompt-context"`. This is rare and
discouraged: put behavioral content in a rule or an
[instruction](https://docs.neuralisapp.com/package-system/instructions.md) and leave reference in plain docs.
