Disclosed weakness · no CVE recorded

GitHub Agentic Workflows setup reused predictable diagnostic paths

GitHub · GitHub Agentic Workflows · setup scripts

First documented disclosure · Review checkpoint

Summary

Setup diagnostics used predictable temporary paths that another local process could pre-create or interfere with.

Severity and exploitation evidence

No CVSS score recorded in this review.

Exploitation not established: The reviewed sources describe a correction or controlled reproduction; malicious exploitation against users is not established.

Exploitation evidence source

Technical severity is not a measure of observed attacks or of risk to your deployment. Attempts, demonstrations, and successful compromise retain the qualifications in the source.

How the weakness works

Diagnostic output paths were not isolated in a private temporary directory.

Conditions required

Another process can influence the relevant temporary paths on the workflow runner.

Potential impact

Local interference with diagnostic output or file writes was possible within runner permissions; remote code execution is not established.

Affected versions and configurations

The reviewed upstream report does not establish a complete affected-release range.

Fix and mitigation status

1.3.1 contains the correction and is marked prerelease. A corresponding stable release is unverified.

Practical next steps

Use the reviewed 1.3.1 prerelease or the specific setup-script correction while following the operator’s release policy.

Operating lesson

Create temporary diagnostic files inside a private directory owned by the operation.

Evidence and limits

The original issue lists automated alerts; the subsequent PR confirms predictable diagnostic paths but does not confirm the separate command-injection alerts. Tagged verification: setup creates a private mktemp directory and cleans it at exit.

Sources