Skip to content

For the complete documentation index, see llms.txt.

How Promptless learns your docs

Before Promptless creates its first suggestion, it analyzes your existing docs to learn your product, your writing style, and how your docs are organized. It keeps learning from your edits and feedback. For rules your existing pages don’t show, add an AGENTS.md file to your docs repository. Promptless also uses the Agent Skills you keep there.

  • Your first suggestions need to match the terms, tone, and formatting of your existing pages. Promptless learns them from those pages with Voice Match.
  • A change to one feature affects several pages, such as a “Workspaces” change that touches the “Permissions” and “Invitations” pages. The product ontology tells Promptless which pages to update together.
  • Your team follows a rule your pages don’t make obvious, such as “write workspace, not project.” State it in AGENTS.md, and Promptless follows it in every suggestion.
  • Your team already keeps a skill for a task, such as writing changelog entries. Promptless runs your existing skills when a documentation task matches one.

Promptless builds a searchable index of your docs and maps the relationships between pages, sections, and topics. When it analyzes a trigger, it uses the index to find the relevant content.

Promptless builds a model of your product’s structure: the features you have, how they relate to each other, and the terminology your team uses. This product ontology helps Promptless:

  • Know which pages are relevant when a specific feature changes
  • Understand your product’s concepts and terminology
  • Find connected features that need coordinated documentation updates

For example, suppose your product has a “Workspaces” feature with sub-features such as “Permissions” and “Invitations.” When workspace behavior changes, Promptless can update every page that covers those relationships.

Promptless analyzes how your documentation is written: its tone, sentence structure, formatting patterns, and terminology choices. This analysis powers Voice Match, so suggestions read like the rest of your docs.

Promptless learns patterns such as:

  • Whether you write formally or conversationally
  • The words and phrases you use for concepts
  • How you organize content, use headings, and format examples
  • Your preferences for capitalization, punctuation, and code formatting

When it drafts content, Promptless also refers to similar existing pages to match their style.

Promptless keeps learning as you use it. When you edit suggestions or give feedback, it applies your preferences to future suggestions. When you publish documentation updates, Promptless re-indexes your docs, so it keeps up as your docs and style change.

Create an AGENTS.md file in the root of your docs repository to give Promptless persistent guidance. Use it for project-specific conventions, terminology, and documentation standards that Promptless follows when it creates suggestions.

When Promptless starts work on your documentation, it checks the root of your docs repository for an AGENTS.md file. When it finds one, it adds the file’s content to its instructions and applies your guidance to every suggestion.

Use your AGENTS.md file for:

  • Terminology preferences, such as the words your team uses for concepts
  • Formatting conventions, such as how to structure code examples, callouts, or headings
  • Content guidelines, such as information to always include or leave out
  • Style rules, such as the voice, tone, or writing patterns specific to your documentation
# Documentation Guidelines
## Terminology
- Use "workspace" not "project" when referring to user containers
- Always capitalize "API" but lowercase "endpoint"
- Prefer "authentication" over "auth" in prose
## Formatting
- Use admonitions for warnings and important notes
- Include a "Before you begin" section in procedural docs
- Code examples should include comments explaining each step
## Style
- Write in second person ("you") rather than third person
- Keep sentences under 25 words when possible
- Use active voice

If your repository already has a CLAUDE.md file for Claude Code, Promptless uses that file. Promptless loads AGENTS.md only when the repository has no CLAUDE.md, so an existing Claude Code setup keeps working.

If you keep Agent Skills in your documentation repository, Promptless uses them with no extra configuration. It picks up skills under .claude/skills/, .agents/skills/, and .cursor/skills/. When a documentation task matches a skill’s description, Promptless runs that skill.

The skills you wrote for other AI coding agents apply to your documentation work too. Keep them where they already live.

Promptless scans for skills before each documentation task. A skill is a directory that contains a SKILL.md file. Promptless checks the source directories in a fixed order: .claude/skills/, then .agents/skills/, then .cursor/skills/. When the same skill name appears in more than one of these directories, the first match wins. For example, a skill under .claude/skills/ takes precedence over a same-named skill under .cursor/skills/. Skill names are case-sensitive.

To shape when and how Promptless uses your skills, add guidance to the PROMPTLESS.md file in your Knowledge Base. For example, tell Promptless to prefer a particular skill for changelog entries, or to run a formatting skill before it publishes API reference pages. Promptless reads this guidance alongside your skills when it decides which skill fits a task.