Quick answers
- Is CVE-2026-123796 exploited in the wild?
- No. Junglewise Threat Intelligence has no report of CVE-2026-123796 being exploited in the wild.
- How severe is CVE-2026-123796?
- Medium severity, with a CVSS score of 4, published 9 November 2026.
- Which products and versions are affected?
- tmatens/compose-lint (GitHub Actions) = 0.32.0, ECOSYSTEM:introduced=0.32.0,fixed=0.33.0|GIT:introduced=2d42617e4c6416f797bbfc6a057950131fa02df3,fixed=691e08bd1881ada886e3f556d0793079e8114451; compose-lint (PyPI) = 0.32.0, ECOSYSTEM:introduced=0.32.0,fixed=0.33.0; Tmatens Compose-Lint 0.32.0 to 0.33.0.
- Is there a fix for CVE-2026-123796?
- Yes. tmatens/compose-lint (GitHub Actions): upgrade to 0.33.0 or later; compose-lint (PyPI): upgrade to 0.33.0 or later. Until then: Use the --allow-partial-coverage flag if upgrading is not immediately possible, though this does not prevent secret disclosure; the advisory recommends upgrading to 0.33.0 as the proper fix. Junglewise finds every place you run it, drafts the upgrade and applies it once you approve.
- How is CVE-2026-123796 attacked?
- It is attacked over the network, in a software library.
- Am I affected by CVE-2026-123796?
- Check whether you run tmatens/compose-lint (GitHub Actions), compose-lint (PyPI) or Tmatens Compose-Lint at an affected version. Junglewise does that check for you around the clock: connect your code, cloud and domains, and it matches new vulnerabilities like this one against what you actually run and tells you when one reaches you.
How to fix
- tmatens/compose-lint (GitHub Actions): upgrade to 0.33.0 or later
- compose-lint (PyPI): upgrade to 0.33.0 or later
pip install "compose-lint>=0.33.0"
Use the --allow-partial-coverage flag if upgrading is not immediately possible, though this does not prevent secret disclosure; the advisory recommends upgrading to 0.33.0 as the proper fix.
Technical details
compose-lint's 0.32.0 introduced a remediation to replace resolved .env values with ${NAME} references in findings output. However, the finding-limit early-exit path raises FindingLimitError before the _quote_references step completes, causing already-graded findings to emit with unredacted secret values in text, JSON, SARIF, and Code Scanning formats. Exploitation requires a workflow that places secrets in a sibling .env file and lints untrusted changes; the condition is triggered when a malicious document combines secret-referencing fields with aliased content exceeding 20,000 findings.
Affected products
- tmatens compose-lint 0.32.0 to 0.33.0
Identifiers
- CVE-2026-123796
- GHSA-rm8w-w798-g545
Timeline
- 2026-11-09: disclosed
- 2026-10-04: patched