SERVICES · LAS VEGAS NV

Charles Trinh — TLS and OAuth Hardening | Las Vegas, NV · Las Vegas, Nevada · Henderson · Clark County, Nevada · North Las Vegas, Nevada

Charles Trinh — TLS and OAuth Hardening, Las Vegas NV
Charles Trinh — TLS and OAuth Hardening

Hardening transport and delegation: certificate lifecycle, cipher policy, redirect URI exactness, token audience and the four OAuth failures that actually get exploited.

Certificates are the most common outage

Expiry takes a surface down with no application error in any log. Monitor expiry from outside the estate and automate renewal, then verify the automation by letting it renew something.

Chain completeness breaks more clients than cipher negotiation does, so test with the real clients rather than a browser tab.

Redirect URIs, exactly

Match redirect URIs byte for byte, including trailing slashes and query components, and reject anything that is not identical to a registered value.

Normalising before comparison is the subtle defect: a server that lowercases or URL-decodes before matching accepts a URI that differs from the one registered.

Audience and scope reach the data

Bind token audience at issuance from client configuration, never from a request parameter, or the client chooses its own scope of access.

Enforce scopes where the data is fetched. A check in the route handler plus a fetch in a shared service means a changed parameter reads a different patient.

Rotation and reuse detection

Rotate refresh tokens on every use and revoke the whole family when a previously used token reappears. That detection is the only signal you get that a token store is compromised.

Log the family identifier and the revocation reason, because it fires exactly once and you need to know which patient, which app and which grant.

The certificate inventory

Every certificate on the public surface belongs in an inventory with its hostname, issuer, expiry, owner and renewal mechanism. The inventory is what turns expiry from a surprise into a schedule, and it is the artefact that shows, before an incident, which renewals are automated and which are manual. Certificates issued by hand are the ones that expire during a holiday, because the person who renews them is the mechanism.

Where automation exists, verify it by allowing a renewal to happen rather than by reading the configuration. Automation that has never run is an intention, and the failure mode is a successful test in staging and a lapse in production where the challenge path differs. The verification costs an hour once a year and it is the difference between a control and a hope.

Cipher and protocol policy

The policy is a decision, not a default, and it should state the minimum protocol version, the permitted cipher suites and the reason for the floor. Recording the reason matters because the next engineer will find a client that fails and will lower the floor to fix it, which is sometimes right and should be a visible decision rather than a quiet one. Where an old client must be supported, name it and date the support rather than accommodating it indefinitely.

Test the policy from outside the estate with the clients that actually connect, including the ones that are not browsers. Medical devices and integration engines negotiate differently from a modern phone, and the connection that fails in production is usually the one that was never in the test matrix.

Redirect URI discipline

The redirect URI is the place where an authorization code is delivered, which makes it the most sensitive value in a client registration. Match it byte for byte, reject anything that differs, and never normalise before comparison: a server that lowercases, strips a trailing slash or decodes a query before matching accepts a URI that is not the one registered. The defect is invisible in normal operation and exploitable by anyone who reads the client's registration.

Where a client needs multiple URIs, register them explicitly rather than pattern-matching. Wildcards in a redirect URI are a bug with a configuration interface, and every recorded incident involving one has the same shape.

Tokens, audiences and scopes

Bind the audience at issuance from the client's configuration, not from a parameter in the request, because a client that can choose its audience can choose where its token is accepted. Enforce scopes where the data is fetched rather than where the route is declared, since a route that checks a scope and then calls a shared service that does not means a changed path parameter reaches a different patient.

Keep access tokens short-lived and refresh tokens rotating, and revoke the family on reuse. That detection fires at most once per compromise and it is the only reliable signal that a token store has been read by somebody who should not have it.

Logging that answers the incident question

Log authorization decisions with the client, the subject, the scope, the resource and the outcome. The question after a disclosure is which application acted for which person against which record, and a log that records only the route cannot answer it. Keep the subject reference hashed where the log leaves the estate, and retain the log for the period the disclosure obligations require rather than the period the platform finds convenient.

Alert on the anomalies that matter: a client suddenly requesting scopes it never used, a subject authenticating from a new region, and a refresh reuse event. Those three cover most of what a compromised client looks like in practice, and each is cheap to instrument.

Working with counterparties

Every federation partner has its own policy, and the integration that matters is the one that negotiates lowest. Establish the agreed minimum with each partner in writing, including protocol version, accepted cipher families and the key strength floor, and record it where the client registration lives. When a partner asks to lower it, the request is a decision with a date rather than an exception granted by whoever was on the call, and the record shows which partner asked and what was decided.

Test with the partner's real client before go-live, using their certificate chain and their resolver rather than yours. The failures that reach production are almost always chain or trust store differences, and they are cheap to find when both teams are in the room and expensive to find at cutover.

Keep the partner list current. A federation accumulates inactive partners, and an inactive partner with a live client registration is an access path nobody is monitoring.

Documentation for the auditor

Keep a single page per client registration: the client identifier, the redirect URIs, the granted scopes, the audience, the token lifetimes and the date of the last review. That page answers the question an auditor asks about delegation, and it also answers the question the application team asks when a call fails. Two audiences, one artefact, and it takes an hour to create per client.

Review the registrations annually and remove the ones with no traffic. Dormant clients are the most common source of an unexpected access path, and the review is the only mechanism that finds them.

Client secrets are not the control

A public client cannot keep a secret, and a confidential client's secret is a shared credential that leaks through configuration files and logs more often than through theft. The controls that actually hold are short token lifetimes, exact redirect matching and proof of possession where the protocol supports it. Treating the secret as the security boundary produces a design that looks careful and fails to a misconfigured repository.

Scan the codebase and the deployment configuration for secrets as part of the pipeline, and rotate any that appear. The scan finds the ones that were committed by accident, which is the ordinary case and the one that matters.

Proof

  • ONC (g)(10) self-test PASS, 317 of 317 subtests, April 2026 — public suite URL
  • 67 public hostnames, 51 serving, one AWS account
  • 104 clearinghouse enrolments across 53 jurisdictions
  • USCDI v7 proposal — Patient Digital Identity Credential — listed at isp.healthit.gov
  • SAM.gov UEI MVKQW9DKKGB8 · CAGE 20KM5 · NPI 1194645564
  • AWS Business Associate Addendum active since 18 August 2026

Where this work is delivered

Engagements run from Las Vegas and reach across the valley: clients in Las Vegas, NV; Las Vegas, Nevada; Henderson; Clark County, Nevada; and North Las Vegas, Nevada — plus remote delivery nationwide.

How an engagement runs

1. Trace one real case

Follow a single patient or claim end to end with a timestamp at every hop, before any document is written.

2. Written findings

A findings list naming the defect, the system, the owner and the date.

3. Monthly review

Review the list monthly and check whether the dates held.

Frequently asked

What is this service?

This page describes charles trinh — tls and oauth hardening — the scope, the failure modes it prevents and how the work is measured. It sits in the Security and Compliance category at trinh.io.

Where is the work delivered?

Engagements run from Las Vegas, Nevada across the valley — Las Vegas, Henderson, North Las Vegas, Spring Valley, Paradise, Winchester, Sunrise Manor, Enterprise, Summerlin, Clark County — and remotely anywhere in the United States. On-site work in the Las Vegas valley, remote delivery elsewhere.

What does it cost?

Standard rate $195 per hour, committed $165 per hour at twenty hours or more per week for three months or more; $1,600 standard day, $1,350 committed day. Scoped in writing before work starts, fixed fee where the deliverable is defined.

Who is Charles Trinh?

A technology architect in Las Vegas, Nevada, working in healthcare identity, consent and interoperability. Not a physician, not a medical practice, and not affiliated with any other Charles Trinh.

How does an engagement start?

With one trace of a real case and a written findings list naming the defect, the system, the owner and the date. The first call is twenty minutes and free: 702-559-3489.

Engagement

Scoped in writing before work starts; fixed fee where the deliverable is defined. First call is 20 minutes and free. 702-559-3489 · charles@trinh.io

Talk to Charles

Record

Profile

Healthcare technology architect working where federal regulation becomes production infrastructure. Built and operates a FHIR R4 clinical platform — SMART App Launch authorization server, patient and provider portals, SQLite — on a hardened single-host AWS estate of 67 public hostnames. The platform returned a self-test PASS on the ONC (g)(10) Standardized API test suite in April 2026, 317 of 317 subtests, publicly verifiable. Designed a consent architecture in which a permissioned ledger records patient grant, refusal and revocation as immutable events while all clinical data stays off-chain and encrypted per patient — no PHI on the ledger by design. Filed a proposed USCDI v7 data element with ONC on verifiable patient identity. Translates the Federal Register, HHS/CMS guidance and HITRUST requirements into architecture, controls and audit evidence.

Selected engineering scale

  • Self-managed estate of 67 public hostnames, 51 serving, on one AWS account.
  • FHIR R4 API with a SMART App Launch authorization server; patient and provider portals; an AI service layer; Hyperledger Fabric peer, orderer and CA; IPFS; a 337-table SQLite clinical store.
  • Two permissioned distributed ledgers in production, one operating a consent registry that records grant, refusal and revocation with no PHI on chain.
  • X12 005010 pipelines across eight transaction sets.
  • Designed, built, deployed and operated end to end by one person.

Credentials and standards work

  • Inferno (g)(10) self-test PASS, 317 of 317 subtests, April 2026 — the ONC Standardized API test suite. Standards validated: US Core 6.1.0, USCDI v3, SMART App Launch 2.0.0, Bulk Data 2.0.0. Public suite URL.
  • USCDI v7 public comment submission, April 2026 — proposed Patient Digital Identity Credential under the Provenance data class, listed publicly on the ONC Interoperability Standards Platform.
  • SAM.gov UEI MVKQW9DKKGB8 · CAGE 20KM5 · NPI 1194645564. AWS Business Associate Addendum active 18 August 2026.

Technical environment

AWS us-east-1 (compute, storage, messaging, SMS, email, voice) · Linux and systemd hardening · nginx · Fabric and EVM ledgers · FHIR R4 · X12 005010 · HL7 · SQLite · self-hosted mail, drive, search and monitoring · Git-based change control and CI release gates.

TRINH — Charles Trinh, technology architect
TRINH

Charles Trinh

Fractional CTO/CIO and Technology Architect — Healthcare Data, Identity & Regulated Infrastructure · Las Vegas, NV

Las Vegas, Nevada · US citizen · Fractional and project engagements

(702) 559-3489 Email Charles Trinh