Welcome to Advisors Crypto for Agents.
The SEC-registered policy layer between AI agents and capital. This documentation covers everything you need to build, ship, and operate an agent inside our guardrails.
What this is
Advisors Crypto for Agents is a programmable interface to a regulated investment policy. Your AI — Claude, Cursor, a custom orchestrator, anything that can speak HTTP or MCP — reads state, validates proposed trades against the user’s Investment Policy Statement, and queues actions for human attestation in the AC app. We never expose an execute scope.
Who it’s for
- Builders shipping agentic finance products who want to inherit a regulated wedge instead of building one.
- Researchers running portfolio strategies that should be bounded by a policy contract, not a system prompt.
- RIA-aligned platforms needing a compliant attestation surface for AI-originated trade ideas.
Why there’s no execute scope
Letting an agent execute compresses every risk decision into one moment of model judgment. The policy layer hoists those decisions out of the prompt and into a contract you defined off-line. Proposals get dry-runnable, audit-logged, and gated by a human attestation step that the agent cannot bypass.
Where to go next
Pick the right entry point for your goal: