Build agents. Automate the business.
Put it on Record.
Build agents and business-process automations from any IDE — or let your coding agent build them for you with the Record skill. Deploy in one command, and every action ships governed: policy, JIT credentials the agent never holds, human approvals, and audit, built in.
Claude Code
Cursor
Windsurf
Codex
Gemini CLIBuild agents from your editor — or have your coding agent build them.
Record ships as a skill your coding agent loads. Stay in the tools you already use — describe the automation, and Record turns intent into a governed, deployed agent.
Any IDE, any coding agent
Claude Code, Cursor, Windsurf, Codex, Gemini CLI — install the Record skill once, and the coding agent you already use learns to scaffold, configure, deploy, and manage governed agents for you.
Describe the automation
Tell your coding agent what the business process should do. It writes the agent config, the Cedar policy, and the tools, and wires the connections — you review the diff, not boilerplate.
Deploy governed by default
One command ships the agent to production with policy, JIT credentials the agent never holds, human approvals, and audit already on. Trust and security are not a later phase.
Build any agent. Any pattern.
Autonomous reasoning, sequential pipelines, parallel fan-out, iterative loops — compose sub-agents, add tools, deploy. All governed automatically.
LLM-driven reasoning with tool use
Deterministic pipeline of sub-agents
Fan-out, execute, aggregate
Iterate until condition met
You build the automation. Record handles the trust.
Identity, credentials, policy, approvals, audit, and spend are built in and on by default. You ship business logic; every action is governed, attributed, and on the record — no security plumbing to wire. Every action can answer: which agent acted, on whose behalf, under which policy, with which credential — and what actually happened.
Stop building plumbing. Start building agents.
Record Gateway, Record Runtime, Record Brain, policy, incidents, diagnostics, observability, credentials — all built in as platform contracts. Enable what the workflow needs without rebuilding the plumbing.
100+ models. Zero integrations.
OpenAI, Anthropic, Google, Mistral, Cohere, Bedrock — all routed through one gateway. Per-agent model selection, automatic fallback, cost tracking, and content filtering. You never write a provider integration again.
Your Dockerfile. Our guardrails.
Bring the managed runtime dependency set the workflow needs. Record Runtime wraps managed agents with sandbox lifecycle, egress-side credential injection, Cedar policy enforcement, and visibility beyond the tool name.
Agents that remember everything.
6-category persistent memory — profile, preferences, entities, events, cases, patterns. Auto-extracted after every session. Vector search. Survives restarts. Your agents learn and get better over time.
Governance that follows the action.
Managed agent actions pass through runtime hooks, Record Gateway, and sandbox/egress enforcement. Cedar policies, HITL approval, and JIT credentials stay in platform contracts.
Connect anything. Govern everything.
MCP servers, OpenAPI specs, custom Python, and messaging channels become governed capabilities through Record Gateway. Managed agents add the runtime and sandbox gates.
MCP, OpenAPI, Python — unlimited.
Point at an MCP server, drop in an OpenAPI spec, or write a plain Python function. Record auto-generates governed tools with JIT credentials. No SDK, no wrapper, no boilerplate.
Deploy once. Trigger from anywhere.
REST API, webhooks, Slack, Teams, Discord, A2A — channels enter through Record Gateway, and managed agents add runtime and sandbox evidence.
Every gate. Every token. Every dollar.
Gateway traces for connected agents, runtime and kernel-level traces for managed agents, per-agent cost attribution, budget enforcement, and quality signals. All queryable via CLI or API.
Let's prove it works.
No sales deck. No feature demo. We run your agents and measure the results.