Gå til hovedinnhold

Edit and check the content in your project

kapi reads your existing files, resolves the terms and writing guidance that apply, and reports findings from the checks you configure. It is built on the open neokapi content engine.

Start a project or use kapi with your agent.

Start with one language​

A help article and an API reference can describe the same feature with different terms. Bind each collection to its writing guidance, inspect what applies to a file, then edit the content and run the configured checks.

The first-project walkthrough uses an existing repository. You can inspect the file diff and each reported finding before accepting an edit. Supported formats have different preservation properties; see the format evidence for the formats you use.

Understand a check result​

A missing placeholder or a prohibited term can be checked against an explicit rule. A passing gate means those configured checks met their thresholds. Factual accuracy and overall writing quality need their own analysis and review; an absence of rule findings does not establish either.

Inspect the available checks and choose which should block your workflow. People and agents can read the applicable guidance before writing, then use findings to locate and repair specific issues.

Add languages when needed​

Use kapi up to draft changed content, reuse content memory and collect work that needs a person. Review and approve changes in the desktop app, and run configured gates in CI. A project with one source language can use the same project guidance and checks.

Three ways to run it​

kapi is available through the CLI, an agent skill, an MCP server and the desktop app. These surfaces use the same engine, tools, flows, and kapi projects, so a workflow built in one runs in the others:

  • Kapi CLI: a single binary that operates directly on files. Best for automation, scripting, and CI.
  • Your AI agent: the agent skill and the MCP server are how the context reaches whichever AI you already use: Claude, or any skill/MCP-capable tool, driving kapi by plain language inside its guardrails. See Use with Claude.
  • Kapi Desktop: a visual app where a person reads and edits the project's context (voice, terms, memory, coordinates), sees where it stands, and works the checks and the review queue. No YAML, no flags.

Not sure which fits? See Kapi CLI or Kapi Desktop?

How you'll work​

You drive Kapi either ad-hoc (one-off, on files you name) or as a project (a kapi.yaml recipe plus a store that remembers your work): the same commands, a different binding. See Projects.

From there it's task by task: the Kapi CLI guides walk preparing your source and going multilingual, and Kapi Desktop does the same work visually.

Next​

  • Context: the model kapi holds; where a piece of content sits, and what governs that place.
  • Installation: the CLI (Homebrew or download) and Kapi Desktop (macOS cask).
  • Quick Start: do something in your browser, no install.
  • Use Kapi with your AI agent: wire it into Claude, Cursor, and more.