Point it at an API you already have and get a production Agent Client Protocol endpoint with auth, permissions and audit in one command.
ACPlify turns an OpenAPI spec, a database or an internal service into a hosted ACP endpoint with per-tool scopes, approval steps and audit logs. It is built for platform and API teams whose customers want an agent integration.
No account. No card. No sales call.
- A hosted ACP endpoint from an API you already have
- Per-tool scopes, approvals and audit logs
- No protocol code to write
The problem
Every company now has to expose its systems to AI agents, and each one is rebuilding the same plumbing badly: transport, session handling, per-tool permissioning, token scoping, rate limiting, audit logging and human approval steps.
It takes an engineer six weeks and produces a security review nobody can pass.
Who it is for
Platform and API teams
Teams at 20 to 500 person software companies whose customers are asking for an agent integration.
Internal tooling teams in enterprises
Teams told to make internal systems agent-accessible this quarter, who need a security review they can pass.
Generate a production config you can copy
Pick authentication, region, client and limits. Copy a config and client snippet that run, with the latency they imply. Runs in your browser. No account, no card, no call.
# my-service — generated by ACPlify
# Copy to my-service.config.yaml. Every value below is live, not a placeholder.
apiVersion: v1
kind: Service
metadata:
name: my-service
region: us-east
spec:
endpoint: https://my-service.acplify.com/v1
auth:
mode: apikey
credential: ${MY_SERVICE_TOKEN}
limits:
requestsPerMinute: 600
timeoutMs: 10000
maxRetries: 2
backoff: exponential
observability:
structuredLogs: true
tracing: w3c
replayRetentionDays: 7curl -sS https://my-service.acplify.com/v1/run \
-H "X-API-Key: $TOKEN" \
-H "Content-Type: application/json" \
--max-time 10.0 \
-d '{"input": {"task": "hello"}}'import { Client } from "@my-service/sdk";
const client = new Client({
endpoint: "https://my-service.acplify.com/v1",
auth: { mode: "apikey", token: process.env.TOKEN! },
timeoutMs: 10000,
maxRetries: 2,
});
const result = await client.run({ task: "hello" });
console.log(result.data);- API keys do not expire on their own. Set a rotation reminder now.
Take this with you
No email required. It is your result.
ACPlify — config · my-service · us-east · apikey · p50 25ms, p99 80ms · 1,500,000 requests/month at 600 rpm burst Run it yourself: https://acplify.com
How it works
ACPlify, from the first step to the result.
- 1
Import your API
Bring an OpenAPI spec, a database or an internal service.
- 2
Set scopes and approvals
Decide what each tool may do and which actions need a human to approve them.
- 3
Get a hosted endpoint
ACPlify serves a production ACP endpoint with OAuth passthrough and rate limiting in front of your API.
- 4
Review the audit trail
Structured logs and replayable sessions show every call an agent made.
Your existing API, ready for agents
A managed protocol gateway that handles the plumbing every team is rebuilding: transport, sessions, permissions, rate limits and audit.
Import what you already have
Point it at an OpenAPI spec, a database or an existing internal service, and get a hosted ACP endpoint.
Per-tool scopes
Grant each tool its own scope, so an agent can read what it needs without being handed everything.
Keep your auth model
OAuth passthrough means agents act with your users' existing permissions. Nothing about your auth has to change.
Approval-required actions
Mark sensitive actions as needing human approval before an agent can complete them.
Structured audit logs
Every call is written to a structured audit log, which is what a security review actually asks for.
Replayable sessions
Replay any agent session to see exactly what was requested and what came back.
Questions people actually ask
Do we have to write protocol code?
No. You import an OpenAPI spec, a database or an existing service, and ACPlify serves the ACP endpoint. Your team writes no protocol code and keeps its existing auth model.
How are permissions handled?
Each tool gets its own scope, OAuth is passed through so agents act with your users' existing permissions, and sensitive actions can require human approval.
What will a security review see?
Structured audit logs of every call and replayable sessions, alongside per-tool scopes and approval-required actions.
What does it cost?
Free up to 10k tool calls a month. 99 dollars a month covers up to 250k calls, and 499 dollars a month adds SSO, private networking and log export.
Can we self-host it?
Yes. A self-host licence is available at 12k a year for regulated buyers.
Who is it for?
Platform and API teams at 20 to 500 person software companies whose customers want an agent integration, and internal tooling teams inside enterprises that need to make systems agent-accessible.
Why should I trust the config generator?
The config is built from the options you pick, and it lists anything in it that will misbehave under load before you copy it.
Pricing
Priced by tool calls, with a free tier to build on and a self-host licence for regulated buyers.
Free
Enough to build and test your agent surface.
- Hosted ACP endpoint
- Per-tool scopes
- Structured audit logs
Growth
For endpoints in production use.
- Everything in Free
- Up to 250k tool calls a month
Business
For teams with enterprise requirements.
- SSO
- Private networking
- Log export
Self-host
A self-host licence for regulated buyers.
- Run the gateway in your own environment
Prices in USD.
Generate a production config you can copy
Pick authentication, region, client and limits. Copy a config and client snippet that run, with the latency they imply. Runs in your browser. No account, no card, no call.
Open the free toolIt runs in your browser. ACPlify never sees your inputs.