AI Science Tracker

GHSL AI-agent index lists privileged release workflow command injection in unified-github-workflows

ipdxco/unified-github-workflows reusable release-check.yml · AI involvement attributed to GitHub Security Lab AI agent (system not named in advisory)

Finding and attribution

A pull-request-controlled version string can forge GitHub Actions output records and replace a trusted prefix later interpolated into a sed program, enabling command execution in a privileged pull_request_target job.

Product: ipdxco/unified-github-workflows reusable release-check.yml. AI system: GitHub Security Lab AI agent (system not named in advisory). Team / organization: GitHub Security Lab.

CVE identifiers: CVE-2026-92261.

AI and human contribution (source account): GHSL’s primary AI-agent index lists this advisory among findings discovered with AI agents. The item advisory credits Jaroslav Lobačevski and does not describe an agent’s role, so this record retains index-level attribution only.

Human involvement is only described where the source account establishes it; unknown involvement does not mean none. Vendor confirmation of a flaw does not confirm every claim about AI contribution.

Discovery context: The GHSL report confirms a working proof of concept but does not claim exploitation in the wild.

Remediation status: fix documented. Fixed in v1.0.44 on 2026-09-16. The vulnerable workflow’s intended caller uses pull_request_target with contents:write; fork-run approval policy may affect reachability.

A source-documented fix does not establish installed patches or current safety. Source checking does not independently reproduce the finding or establish observed exploitation.

Disclosed 22 Oct 2026. AI attribution dated 22 Oct 2026. Sources checked 14 Nov 2026; these are separate dates, not a live installation check.

Recorded CVSS: Not available · no assessment established here. Documented technical effects: code execution.

Severity, consequences and exploitation evidence

Case grouping uses the highest recorded CVSS score, with its version-specific severity band. It is descriptive, not a combined or editorial risk index. All recorded assessments remain below; scores from different versions and sources may disagree. CVSS specifications define version-specific scoring and severity bands.

No source-backed CVSS assessment is established in this catalog.

CVE-2026-92261: code execution.
Prerequisites / affected scope: A pull request author can inject output records that overwrite a workflow prefix subsequently interpolated into sed with its e flag, resulting in arbitrary command execution in a privileged pull_request_target job with contents:write..
Exploitation evidence: Exploitation in the wild is not established by the cited description.. Impact source · checked 16 Nov 2026.

Documented fix: Fixed in v1.0.44 on 2026-09-16. The vulnerable workflow’s intended caller uses pull_request_target with contents:write; fork-run approval policy may affect reachability.

Original evidence

Download this source record (JSON)

Open this case in the interactive tracker · Browse all static security cases