entityOS Sentinel · Reward Program
If you find a flaw, we want to hear from you first. Report responsibly, get acknowledged, and earn a reward.
Reward Tiers
Indicative ranges in AUD, paid at entityOS's discretion based on validated impact, exploitability and report quality.
Allowlist / mTLS bypass, RCE, SSI credential forgery, cross-tenant data exposure.
Auth bypass, IAM privilege escalation, certificate validation flaw.
Stored XSS in a sensitive context, IDOR, exploitable logic flaw.
Minor information disclosure, non-sensitive misconfiguration.
Duplicate, theoretical, and out-of-scope reports may receive recognition only. One reward per unique root-cause issue.
In Scope & How It Works
API Gateway, interface / VPC endpoints and the IAM roles behind them.
entityos.io, the developer console and any first-party web application.
You submit a clear, reproducible report through this form.
We confirm receipt and assign a reference, typically within a few business days.
We validate, reproduce and assign a severity tier.
We fix the issue, usually inside a ~90-day disclosure window.
You're paid per the tier and credited (if you wish) once the fix ships.
Architecture & Trust Model
entityOS is a fully managed cloud platform. An app built on it is a thin client — static form (HTML & CSS) and function (JavaScript via the entityOS SDK) running in the visitor's browser — that talks to the entityOS API. The API, not the client, is the security boundary: authentication and authorisation are enforced server-side, per object and field. A lot of what looks exposed on the client is public by design. Please factor this in so your report lands in scope.
These are inherent to the thin-client model. On their own they are not findings.
The app's HTML, CSS and JavaScript — including SDK controllers and list/form configuration such as the util-view-table field definitions — are delivered to the browser and fully inspectable via View Source / DevTools. Reading or modifying client code is expected.
app.dataReturned records are cached client-side in the app.data object for the signed-in session. Seeing your own data there, or in XHR responses, is not a disclosure.
The client is untrusted. Calling the API directly, or editing the JavaScript to add fields or change parameters, cannot exceed the permissions the server grants the authenticated user. "I changed the client" is not, by itself, a finding.
Static files served at site<nnnn>.entityos.cloud are public to the app's users by design.
Hosting, servers and OS / platform patching are managed by entityOS / ibCom. Generic version banners or platform-owned components are usually out of scope.
The test is always the server-side API response, not the client.
A direct API call or modified client that returns or writes data the signed-in user should not have access to.
One space (account) reading, writing or enumerating another space's data.
OBJECT_SEARCH / OBJECT_MANAGE calls returning or mutating records outside the user's grant by changing an id or parameter.
Bypassing logon, reusing or forging a session, or gaining rights the account was not granted.
Input the server-side API acts on rather than rejecting, or a workflow that can be driven to an unintended state.
Everything listed under In Scope — the closed-internet edge, mTLS/PKI, self-sovereign identity, and the API / interface endpoints.
When you report, tell us what the server-side API did — the request you sent and the response you got back. A finding that reduces to "the client exposes X" or "I can edit the JavaScript" is solved by architectural design and will be closed as out of scope.
Submit a Report
One issue per submission. Include enough detail for us to reproduce it. Don't access or modify data that isn't yours, and don't run denial-of-service or spam tests.
Thank you for disclosing responsibly. We've logged your report and the team has been alerted. Use the following reference in the subject, and the password you set to zip any evidence and send to security@ibCom.biz:
We'll acknowledge by email shortly. Please keep the issue confidential until it's resolved.
Safe Harbour
If you make a genuine effort to follow this policy, we will treat your research as authorised, will not pursue legal action, and will work with you to understand and resolve the issue quickly.
In short: report only what you find, don't pivot or exfiltrate, give us reasonable time to fix it, and acknowledge it.