CVE-2026-32245 is a medium-severity (CVSS 6.5) CWE-863 vulnerability in github.com/steveiliop56/tinyauth. A fix is available for github.com/steveiliop56/tinyauth — see the affected versions and patch details below.
Tinyauth's OIDC authorization codes are not bound to client on token exchange
Exploitation Status
Proof-of-concept exploit code exists
- CISA’s SSVC triage found public proof-of-concept exploit code for this CVE, though no confirmed active exploitation.
Exploitation and automatability from CISA’s SSVC triage for CVE-2026-32245.
EPSS Exploitation Probability
EPSS (Exploit Prediction Scoring System) is a daily probability model maintained by FIRST.org. It estimates the likelihood a CVE will be exploited in production environments within the next 30 days, derived from real-world threat intelligence signals.
How urgent is this, really
CVE-2026-32245 plotted by exploitation likelihood (EPSS) against impact (CVSS). The shaded corner — EPSS 50%+ and CVSS 7.0+ — is where this CVE doesn't sit, though severity or exploitability alone can still warrant action.
Where this sits among everything scored
Of 377,333 CVEs with a current EPSS score, this one falls in the < 10% band (highlighted). Real counts from FIRST.org, not a sample — log-scaled since the landscape is heavily right-skewed.
Real-World Exposure
github.com/steveiliop56/tinyauthReal-time download stats are indexed for npm and PyPI packages. This vulnerability affects Go packages — download data is not available via public APIs for these ecosystems.
Description
Summary
The OIDC token endpoint does not verify that the client exchanging an authorization code is the same client the code was issued to. A malicious OIDC client operator can exchange another client's authorization code using their own client credentials, obtaining tokens for users who never authorized their application. This violates RFC 6749 Section 4.1.3.
Details
When an authorization code is created, StoreCode at internal/service/oidc_service.go:305-322 correctly stores the ClientID alongside the code hash in the database (line 316).
During token exchange at internal/controller/oidc_controller.go:267-309, the handler retrieves the code entry at line 268 and validates the redirect_uri at line 291, but never compares entry.ClientID against the requesting client's ID (creds.ClientID). The code proceeds directly to GenerateAccessToken at line 299.
The developers clearly intended this check to exist, the refresh token flow at internal/service/oidc_service.go:508-510 has the exact guard: if entry.ClientID != reqClientId { return TokenResponse{}, ErrInvalidClient }. It was simply omitted from the authorization code grant.
The entry.ClientID field is stored in the database but never read during authorization code exchange.
PoC
Prerequisites: a tinyauth instance with two OIDC clients configured (Client A and Client B). Both clients must have at least one overlapping redirect URI, or the attacker must be able to intercept the authorization code from Client A's redirect (via referrer leak, browser history, log access, etc.).
Step 1 — Log in as a normal user:
curl -c cookies.txt -X POST http://localhost:3000/api/user/login \
-H "Content-Type: application/json" \
-d '{"username":"admin","password":"admin123"}'
Step 2 — Authorize with Client A:
curl -b cookies.txt -X POST http://localhost:3000/api/oidc/authorize \
-H "Content-Type: application/json" \
-d '{"client_id":"client-a-id","redirect_uri":"http://localhost:8080/callback","response_type":"code","scope":"openid","state":"test"}'
Extract the code parameter from the redirect_uri in the response.
Step 3 — Exchange Client A's code using Client B's credentials:
curl -X POST http://localhost:3000/api/oidc/token \
-u "client-b-id:client-b-secret" \
-d "grant_type=authorization_code&code=<CODE_FROM_STEP_2>&redirect_uri=http://localhost:8080/callback"
The server returns a valid access_token, id_token, and refresh_token. Client B has obtained tokens for a user who only authorized Client A.
Impact
A malicious OIDC relying party operator who can intercept or observe an authorization code issued to a different client can exchange it for tokens under their own client identity. This enables user impersonation across OIDC clients on the same tinyauth instance. The attack requires a multi-client deployment and a way to obtain the victim client's authorization code (which is passed as a URL query parameter and can leak through referrer headers, browser history, or server logs). Single-client deployments are not affected.
Affected Packages
| Ecosystem | Package | Vulnerable range | Fix |
|---|---|---|---|
| 🐹Go | github.com/steveiliop56/tinyauth | all versions | 1.0.1-20260311144920-9eb2d33064b7go get github.com/steveiliop56/tinyauth@v1.0.1-20260311144920-9eb2d33064b7 |
Detection & mitigation playbook
Open-source dependencyDetect
Scan your dependency tree (package-lock.json, pnpm-lock.yaml, requirements.txt, go.sum, etc.) for github.com/steveiliop56/tinyauth, including transitive dependencies — a direct dependency you never call can still pull in a vulnerable version.
Fix
Update github.com/steveiliop56/tinyauth to 1.0.1-20260311144920-9eb2d33064b7 or later, then make sure no transitive (indirect) dependency still pins the vulnerable range — O3 confirms CVE-2026-32245 is resolved across your whole dependency graph.
Workarounds
If you can't upgrade right away: gate or disable the affected feature, validate untrusted input at the boundary, and avoid passing attacker-controlled data into the vulnerable path. O3's runtime protection blocks exploitation in production as an interim safeguard until the upgrade lands.
How O3 protects you
O3 Security's impact-aware SCA analyses which vulnerable code paths your application actually calls, so a match like CVE-2026-32245 can be triaged on real exposure rather than presence alone.
Tailored to CVE-2026-32245. Runtime protection reduces exposure until a permanent patch is applied and verified — it complements patching, it doesn't replace it.
Frequently Asked Questions
Is CVE-2026-32245 in your dependencies?
O3 Security finds CVE-2026-32245 across Go dependencies, including transitive ones, and its impact-aware SCA ranks findings by whether your code actually calls the vulnerable path.