Connect AgentLed to your AI agent
Run one command. Then talk to your agent.
npx @agentled/cli setup
Your AI agent (Claude Code, Codex, Cursor, Claude Desktop, Windsurf) is now connected to AgentLed. Restart it, then describe what you want.
Implementation detail (auth, folder scaffold, MCP config, skill install) lives in /docs/cli.md.
What you can ask your agent
Plain language. The agent has 100+ integrations, persistent memory, and a workflow design playbook.
Lead generation
"Find fintech CTOs in Europe, enrich via LinkedIn + Hunter, score by ICP fit, draft personalized outreach, save to Knowledge Graph."
Inbound qualification
"Qualify form submissions daily, score against my ICP, and email me the top 10 with a one-line summary each."
Investor matching
"Match this startup against our investor database. Score by sector focus, stage, and check size. Compare with last round."
Content & competitive intel
"Pull competitor pricing pages weekly, summarize changes, and post a diff report to Slack with a public read-only link."
Workflows compound
One workflow's output is the next workflow's context. The Knowledge Graph is the substrate — every agent and every workflow in your workspace reads from and writes to it.
Example chain:
- Sourcing workflow finds fintech CTOs in Europe and writes them to
kg.list.leadswithstatus: "new". - Outreach workflow tests three message variants on a small sample and notes which performed.
- A learning agent records the outcome in
knowledge.icp.fintech-cto-eu— "variant B (problem-first) got 18% reply rate vs. 6% for variant A." - Content workflow for next week's blog post pulls from the same ICP record — angles aligned with what already converted.
- A new agent joining the workspace inherits all of this on day one. No retraining, no rebuilding.
The point isn't any single workflow — it's that every run sharpens the substrate everything else depends on. ICP, tone, products, what worked, what didn't — your agents share one brain.
What your agent gets
A focused toolset, not a library to memorize.
| Capability | Tools |
|---|---|
| Build & run workflows | create, validate, publish, start, stop, retry |
| Knowledge graph | read/write rows, traverse edges, list scoring history |
| Test before deploying | test AI / app / code actions, capture fixtures, replay zero-credit |
| Persistent memory | store / recall / search across sessions and runs |
| Integrations | 100+ apps via unified credits — LinkedIn, Hunter, OpenAI, Anthropic, Apify, and more |
Two layers of memory
Other tools start from zero every run. AgentLed maintains two persistent memory layers:
Business-level
Company context shared across all workflows and users. ICP definitions, scoring models, client profiles, workflow outcomes. One workflow learns it — every workflow benefits.
User-level
Individual preferences, personal workflow history, and conversation patterns. Each team member gets personalized automation within the shared business context.
Pricing
Pro
€24–45/mo
For individuals and small teams getting started with AI workflows.
Teams
€87–118/mo
For teams running production workflows with Knowledge Graph and collaboration.
Enterprise
Custom
White-label, on-premise, dedicated support, custom integrations.
Running headless?
AI agents without browser access, CI runners, and remote shells: see /docs/cli.md for the API-key + headless install path.