Preloop vs Helicone: Migrate to an actively developed control plane

Preloop is the open-source AI agent control plane — an AI model gateway plus an MCP firewall, per-call human approvals, runtime observability, and a tamper-evident audit trail, all Apache 2.0 and self-hostable. Helicone is a respected open-source LLM gateway and observability tool. The honest one-liner: if you valued Helicone's open gateway but need an actively developed control plane that also governs tool calls and human approvals, migrate to Preloop.

At a glance

What Helicone does well

Helicone is a well-regarded, widely adopted Apache-2.0 LLM gateway and observability product — reported to serve on the order of 16,000 organizations. It is, architecturally, the closest player to Preloop's gateway.

• Mature LLM observability. Logging, cost tracking, and analytics over model traffic are its strength and the reason many teams adopted it.
• Open source and gateway-first. Apache-2.0 and self-hostable, with a clean OpenAI-compatible proxy.
• Large install base. A lot of teams already route model traffic through it.

Helicone earned its adoption. The reason to look at a migration is not quality — it is that, since the Mintlify acquisition in March 2026, it has been widely reported to be in maintenance mode with no active roadmap, while the agent-governance problem has moved past pure model-traffic observability.

Where Preloop is different

Preloop covers the same model-gateway-plus-observability ground Helicone is known for, and then adds the layers an agent control plane needs — under active development.

• The same open, self-hostable gateway, still moving. Preloop's core is Apache 2.0 with an OpenAI/Anthropic-compatible model gateway, budgets, and per-agent attribution — and it is actively developed rather than frozen.
• An MCP firewall Helicone does not have. Preloop intercepts every tool call and applies allow / deny / require-approval / require-justification policy on the arguments, not just the tool name. Helicone governs model traffic; it does not govern what an agent does with tools.
• Per-call human approvals. When a policy routes a tool call for review, a person decides on mobile, Apple Watch, Slack, Mattermost, email, or webhook — with the tool, arguments, agent reasoning, and matched policy in front of them. This primitive does not exist in Helicone.
• One runtime timeline for tools and models. Runtime session observability shows tool calls, model calls, policy decisions, approvals, spend, and outcomes in one place — a superset of model-only observability.
• Audit as evidence. A tamper-evident, append-only ledger of decisions, approvers, justifications, matched policy versions, and outcomes — usable for security review and EU AI Act readiness work.
• preloop agents discover. One command enrolls Claude Code, Codex CLI, Cursor, Gemini CLI, OpenClaw, and OpenCode on a developer's machine.

Migrating from Helicone

Helicone is OpenAI-compatible and gateway-first, which makes the migration mechanical for the part you already use:

1. Repoint the gateway. Swap your Helicone gateway base URL for Preloop's OpenAI/Anthropic-compatible model gateway. Model traffic, budgets, and per-agent attribution keep working.
2. Keep your observability. Preloop's runtime sessions cover model-call logging and cost tracking, so you do not lose the visibility Helicone gave you.
3. Add the layers Helicone never had. Turn on the MCP firewall for tool-call policy and wire human approvals to the channels your team reads.
4. Onboard local agents. Run preloop agents discover to enroll Claude Code, Codex CLI, Cursor, Gemini CLI, OpenClaw, and OpenCode without SDK changes.

You can run the two side-by-side during the cutover and decommission the frozen gateway once Preloop is carrying traffic.

When to stay on Helicone

• You only need model-traffic logging and cost analytics, have no tool-call governance or approval requirement, and are comfortable with a tool that is reported to be in maintenance mode.
• Your current Helicone deployment meets your needs and you have no near-term agent-governance roadmap.

When to choose Preloop

• Teams on Helicone who need tool-call governance and human approvals that a model-traffic gateway cannot provide.
• Platform and DevEx teams rolling out Claude Code, Codex CLI, Cursor, Gemini CLI, OpenClaw, or OpenCode who need policy and approvals, not just observability.
• Security and compliance teams that need a tamper-evident audit trail and AI Act readiness evidence.
• Open-source-first teams that want an actively developed, self-hostable control plane to standardize on rather than a frozen point tool.

FAQ

Is Helicone still being developed? Helicone is widely reported to be in maintenance mode following the Mintlify acquisition in March 2026, with no active roadmap. It remains usable and open-source, but new capabilities are not expected.

Is Preloop a drop-in replacement for Helicone's gateway? For the model-gateway use case, largely yes. Preloop's model gateway is OpenAI/Anthropic-compatible, so repointing the base URL keeps model traffic, budgets, and attribution working — and you gain the MCP firewall, approvals, and audit Helicone does not have.

Does Preloop keep the observability I relied on in Helicone? Yes. Preloop captures model-call logging and cost tracking in its runtime sessions view, and adds tool calls, policy decisions, and approvals to the same timeline.

Is Preloop open source like Helicone? Yes. Both are Apache 2.0. The difference is scope and momentum: Preloop is a full, actively developed control plane, not a model-traffic gateway in maintenance mode.

How do I onboard existing agents to Preloop? Run preloop agents discover. Preloop finds local Claude Code, Codex CLI, Cursor, Gemini CLI, OpenClaw, and OpenCode configurations, backs them up, and rewrites them to route through Preloop.

Try Preloop

• Start your free trial
• Book a demo
• Preloop on GitHub