Your Buron marketing data and knowledge, from the terminal and CI. Also the fastest way to wire Buron's MCP server and skills into your AI editor.
npm i -g @buron/cli
# or
pnpm add -g @buron/cliRequires Node.js 18+.
buron login # opens the browser; ends linked to your team
buron setup # connect your editors: MCP server + skillssetup writes each editor's MCP config in its own convention (Claude Code .mcp.json, Cursor .cursor/mcp.json, VS Code .vscode/mcp.json, Codex ~/.codex/config.toml) and installs Buron's skills through npx skills add buron-ai/skills, the open agent-skills installer.
Then ask your editor: "What was my ad spend in the last 7 days?"
buron datasets list
buron datasets query cross-channel-ads -m cost,conversions -d campaign_name \
--from 2026-07-01 --to 2026-07-31 --sort cost:desc --limit 10
# Live Google Ads config reads the datasets don't cover
buron gaql "SELECT campaign.name, campaign.bidding_strategy_type FROM campaign WHERE campaign.status = 'ENABLED'"queries and dashboards manage saved queries and run dashboards. Add --json to any read for pipeable output.
buron file read /wiki/company.md
echo "# Q2 recap" | buron file write /wiki/analyses/q2-recap.md
buron file grep "pricing" --directory /wiki/Mint a token, store it as a secret, and every command works headless:
buron token create
gh secret set BURON_TOKEN- run: npm i -g @buron/cli
- run: buron file write /snapshots/$(date +%Y-%m-%d).md --from-file CHANGELOG.md
env:
BURON_TOKEN: ${{ secrets.BURON_TOKEN }}MIT