CLI for interacting with the Plain API.

Specifically for indexing content as Knowledge Sources in your Plain workspaces, to be used by Plain Agents.

• Installation
• Authentication
• Commands
  • Knowledge Sources
  • index-url
  • index-sitemap
• Running in CI

Requires Node.js 22.12 or later.

npm install -g @team-plain/cli
plain --help

To authenticate with your Plain workspace, the CLI uses a Plain API Key, read from the PLAIN_API_KEY environment variable.

export PLAIN_API_KEY=plainApiKey_xxx

Commands for indexing content as Knowledge Sources in your Plain workspace, so that Plain Agents, such as Ari, can use it to answer customer questions and as context.

Each command fetches and indexes the content of the provided URL or sitemap. Running a command again for a URL or sitemap that's already a knowledge source reindexes it.

Index a single page by URL.

Required permissions: knowledgeSource:create

PLAIN_API_KEY=plainApiKey_xxx plain index-url <url>

Options:

• -l, --labelTypeIds <labelTypeIds...>: Array of label type IDs to associate with the indexed url

Index all the urls in a given sitemap you provide.

Required permissions: knowledgeSource:create

PLAIN_API_KEY=plainApiKey_xxx plain index-sitemap <sitemap url>

Options:

• -l, --labelTypeIds <labelTypeIds...>: Array of label type IDs to associate with the indexed urls from the sitemap

Use the CLI in your CI pipeline to index your content as Knowledge Sources, for example when you publish new documentation to a Mintlify site.

- uses: actions/setup-node@v7
  with:
    node-version: 24

- name: Index docs
  run: npx --yes @team-plain/cli@1 index-sitemap https://docs.example.com/sitemap.xml
  env:
    PLAIN_API_KEY: ${{ secrets.PLAIN_API_KEY }}

Greptile auto-approves a PR it scores 5/5 and rates Low risk. Anything under .greptile/ always needs a human, and so does a release: the Version Packages PR is opened by a bot, and Greptile does not review bot PRs.