Check out our open source tools to give Claude Code GTM superpowers
All open-source tools

AI Ark CLI

Orchestrate AI Ark workflows from Claude Code.

Open source · MIT · Production-tested inside LeadGrow's live pipeline
Quick install
$ bun install && bun run build
What you unlock

CLI bindings for AI Ark — trigger workflows, query state, push and pull data. Designed for Claude Code pipelines that treat AI Ark as the orchestration layer for multi-step GTM work.

OutcomeBuilt at LeadGrow, used in live pipelines, shipped open source so you don't have to build it yourself.
How it works
AI Ark CLI — how it works
See it in action

2 things you can do right now.

No UI. No clicking. Just commands that execute.

Trigger a research workflow and wait
$ ark workflow trigger \ --id "company-research-v2" \ --input '{"domain": "notion.com"}' \ --wait
✓ Workflow started (run_01jk3...) ✓ Completed in 23s
OutcomeDeep company research in 23 seconds. Hands off the keyboard entirely.
Pull structured output and pipe downstream
$ ark run output --id "run_01jk3" \ | jq '.outputs' \ | attio people upsert
✓ Research output routed → 1 record updated in Attio
OutcomeWorkflow output lands exactly where it needs to go. Zero manual mapping.
What's included

Every capability, ready to script.

Workflow trigger + status commands
Bidirectional data sync
Claude Code skill-ready
The workflow

End to end. Zero manual steps.

This is how LeadGrow runs research at scale — AI Ark for the heavy lifting, Claude Code to pipe results everywhere they need to go.

01

Trigger the workflow.

Pass a domain or company name. AI Ark runs the full research pipeline — web search, tech fingerprinting, signal extraction — without any manual steps.

$ ark workflow trigger --id "company-research-v2" \ --input '{"domain": "notion.com"}' --wait
02

Check status or block for completion.

Poll for completion with --wait to block inline, or check async. Either way you get structured JSON output when the workflow finishes.

$ ark run status --id "run_01jk3" --format json
03

Push results to your stack.

Pipe structured output to Attio, Bison, or any downstream tool. The CLI handles the data transform so you never touch a spreadsheet.

$ ark run output --id "run_01jk3" \ | jq '.outputs' | attio people upsert
Real scenarios

What teams actually use it for.

Not theoretical. These are the pipelines running at LeadGrow and client stacks today.

01Automating enrichment and research pipelines
02Running AI Ark workflows without touching the UI
Included skills

Pre-built automations. Ready to run.

These aren't demos. They're the Claude Code skills we run inside LeadGrow, shipped open source so you don't have to build them yourself.

research-runner

Company Research Runner

Triggers AI Ark research on any domain. Returns structured intel: tech stack, headcount, recent news, hiring signals. Logs to Attio automatically.

Triggers: "research [company]" or before any discovery call
workflow-trigger

Workflow Trigger

Exposes any AI Ark workflow as a Claude Code command. Say what you want to run. The skill maps it to the right workflow ID and fires it.

Triggers: Any workflow you've built in AI Ark
result-parser

Result Parser

Reads raw AI Ark output and routes fields to the right downstream tools — Attio for CRM fields, Bison for email variables. Zero manual mapping.

Triggers: After any AI Ark workflow completes
Custom build

Want us to build yours?

Custom CLIs, MCP servers, and Claude Code skills wired to your exact GTM stack.

We build these for clients. Then we ship them open.

Book a Strategy Call →