Cybersecurity researchers at Wiz have disclosed a GitHub Actions workflow injection vulnerability in Snowflake’s public snowflakedb/snowflake-connector-net repository. This vulnerability could be exploited through a crafted GitHub issue to execute commands in a workflow that contains internal Jira credentials.
The issue was identified in the jira_issue.yml file, which is triggered when a public issue is opened. It exposed sensitive information such as JIRA_BASE_URL, JIRA_USER_EMAIL, and JIRA_API_TOKEN within the same workflow step. Notably, this vulnerability was limited to the repository’s CI/CD automation, with no affected releases of the Snowflake Connector for .NET identified.
The workflow directly inserted attacker-controlled issue title and body values into a shell run block. It also incorrectly checked the github.event.pull_request.user.login property, which does not exist in the context of an issue event. According to GitHub documentation, dereferencing a nonexistent property evaluates to an empty string, allowing ordinary issues to reach the job.
Wiz reported that their Red Agent system exploited this injection during authorized security testing. After an initial payload resulted in a shell syntax error, the system adapted its approach and successfully obtained the Jira API token used by the workflow.
This token, belonging to qa@snowflake.net, provided read access to Jira projects related to engineering, security compliance, and bug bounty tracking on snowflakecomputing.atlassian.net. The underlying Jira permissions and audit records remain private.
Wiz disclosed the vulnerability to Snowflake via HackerOne on June 23, 2026, under report #3819931. Snowflake merged a fix the same day, replacing direct GitHub expression expansion with environment variables passed to jq as arguments. The vulnerable workflow had been merged into the default branch five days earlier, on June 18.
Snowflake stated that their investigation found no evidence of unauthorized access. The Jira token was rotated on June 24, and Snowflake’s review indicated no unrelated external use during the five-day exposure window. As of August 17, 2026, no CVE or CVSS score had been assigned to this issue, and no malicious exploitation in the wild has been confirmed.
For further details, refer to the report by The Hacker News.
Follow Cyber Warriors Middle East for further global cybersecurity developments.
