# CVE-2026-67354: guzzlehttp/guzzle before 7.15.1 URI Fragment Disclosure via Referer

> Attackers can read sensitive URI fragments from Referer headers if redirect referer logging is enabled, exposing tokens or secrets.

- **Severity:** High 8.2 (`CVSS:4.0/AV:N/AC:H/AT:P/PR:N/UI:N/VC:H/VI:N/VA:N/SC:N/SI:N/SA:N`)
- **Exploited in the wild:** Not known
- **Login needed:** Not needed
- **Attackable:** Over the internet
- **Victim action needed:** Not needed
- **Fix available:** Yes
- **Published:** 1 Aug 2026
- **Updated:** 8 Oct 2026
- **Also known as:** DEBIAN-CVE-2026-67354, GHSA-h95v-h523-3mw8, MINI-32hp-qmpm-vfjj, MINI-53p4-q2mr-557q, MINI-5rwf-pjqh-788c, MINI-7pw7-cxgp-w69w, MINI-gj47-7rcx-f373, MINI-v63c-99m7-hw3p, MINI-w3c3-626x-x45h, MINI-xjrc-8r6q-7f25, UBUNTU-CVE-2026-67354

## What is CVE-2026-67354?

CVE-2026-67354 is a high-severity vulnerability in guzzle, published on 1 Aug 2026. An attacker can exploit it over the internet without logging in, with no action from the victim.

### Official description

guzzlehttp/guzzle versions before 7.15.1 contain an information disclosure vulnerability in RedirectMiddleware. When the optional allow_redirects.referer setting is enabled, the middleware copies the URI fragment (the portion after '#') from the referring request into the generated Referer header when following a same-scheme redirect (e.g., HTTPS to HTTPS). An attacker who controls the redirect destination can read this fragment from the incoming Referer header, potentially disclosing one-time login secrets, access tokens, state values, or other sensitive client data to a server never meant to receive it. The referer setting is disabled by default. Fixed in 7.15.1, which strips the fragment before generating the Referer value.

Weakness: CWE-201 (Insertion of Sensitive Information Into Sent Data)

## Which versions are affected by CVE-2026-67354?

2 open source packages ship an affected version, including guzzlehttp/guzzle.


| Package | Ecosystem | Affected | Fixed in |
| --- | --- | --- | --- |
| `guzzlehttp/guzzle` | Packagist | All versions | 7.15.1 |
| `guzzlehttp/guzzle` | Packagist | All versions | 7.15.1 |

## How do I fix CVE-2026-67354?

Upgrade guzzlehttp/guzzle to 7.15.1 or later.

## Is CVE-2026-67354 being exploited?

There is no known exploitation of CVE-2026-67354 in the wild and no public exploit code.

## Fix prompt for AI coding agents

```text
Check this codebase for CVE-2026-67354 and fix it if it is affected.

What it is: CVE-2026-67354 is a high-severity vulnerability in guzzle, published on 1 Aug 2026. An attacker can exploit it over the internet without logging in, with no action from the victim.
Exploitation: There is no known exploitation of CVE-2026-67354 in the wild and no public exploit code.

Affected packages and fixed versions:
- guzzlehttp/guzzle (Packagist): affected all versions; fixed in 7.15.1
- guzzlehttp/guzzle (Packagist): affected all versions; fixed in 7.15.1

Task:
1. Search the dependency manifests and lockfiles (for example package.json, package-lock.json, requirements.txt, poetry.lock, uv.lock, pom.xml, build.gradle, go.mod, Cargo.lock, Gemfile.lock) and any Dockerfiles for the affected packages, including transitive dependencies.
2. If an affected version is in use, upgrade it to the fixed version for its release line with the project's package manager, so the lockfile is regenerated. Prefer the smallest upgrade that fixes it.
3. If there is no fix, or the upgrade would break the build, do not guess: explain the risk and the safest workaround instead.
4. Run the tests and summarise what changed and why.
5. If nothing in the project is affected, say so and change nothing.

Reference: https://insights.sbomify.com/CVE-2026-67354/ (Markdown: https://insights.sbomify.com/CVE-2026-67354/index.md)
```

## Sources

- CVE record: https://www.cve.org/CVERecord?id=CVE-2026-67354
- NVD: https://nvd.nist.gov/vuln/detail/CVE-2026-67354
- OSV: https://osv.dev/vulnerability/CVE-2026-67354

---
From sbomify Insights: https://insights.sbomify.com/CVE-2026-67354/
Summaries are generated and may contain mistakes; the official description above is the record.