Skip to content
NewA calmer, faster ZeroHour
StoryVulnerabilityhigh

Progress DataDirect GenAI flaw CVE-2026-91140 flagged

3 sources · 3 articlesFirst reported Updated ()3 · 1 read
What's new

First coverage.

Merged summary · grok-4.7 · rewritten as coverage arrivesSummaries by Vani AI

CVE-2026-91140 is a critical command-injection flaw in Progress DataDirect GenAI files before version 2.1; Canada issued AV26-1005 and no active exploitation is reported.

On October 6, 2026, the Canadian Centre for Cyber Security issued advisory AV26-1005 for CVE-2026-91140 in Progress Software's Autonomous REST Connector GenAI component, ARCGenAI-Generator, before version 2.1, pointing to a September 2026 Progress DataDirect critical security alert and urging administrators to apply updates. Cyber Security News and GBHackers, reporting on October 7, describe a critical command-injection flaw in early-access DataDirect Autonomous REST Connector AI Model Generator definitions: a filename taken from an OpenAPI or Swagger document is used in a shell command without sufficient validation, so a crafted document can run arbitrary OS commands. Those outlets name ARCGenAI-Generator.agent.md 2.0, ARCGenAI-Generator.prompt.md 1.0, and ARCGenAI-EntityGen.agent.md 1.0 as affected, with fixes in version 2.1 definitions that customers should pull from GitHub; no installer is required. Progress reports no distinctive error when the flaw is triggered, and GBHackers says Progress published no CVSS score even though coverage calls the issue critical. The Canadian bulletin and GBHackers say there is no evidence of active exploitation, and reported impact is limited to generator workspaces and CI environments. Sources differ in detail: Canada refers broadly to ARCGenAI-Generator before 2.1, while the later reports specify three definition files, and one describes a temporary-file cleanup command while another says the filename is passed into a shell without proper quoting.

Key facts
  • CVE-2026-91140 is a critical command-injection flaw in early-access Progress DataDirect Autonomous REST Connector GenAI agent definitions, including ARCGenAI-Generator before version 2.1.
  • The Canadian Centre for Cyber Security issued advisory AV26-1005 on October 6, 2026, citing Progress's September 2026 DataDirect critical security alert and urging administrators to apply vendor updates.
  • A filename taken from an OpenAPI or Swagger document is passed into a shell command without sufficient validation, allowing a crafted document to run arbitrary OS commands.
  • Named affected files are ARCGenAI-Generator.agent.md 2.0, ARCGenAI-Generator.prompt.md 1.0, and ARCGenAI-EntityGen.agent.md 1.0; corrected definitions are version 2.1 and are pulled from GitHub, with no installer required.
  • The Canadian bulletin and GBHackers report no evidence of active exploitation; GBHackers also says Progress published no CVSS score.
  • Progress reports no distinctive error on exploitation, and described impact is limited to generator or developer workspaces and CI environments.
Analyze CVE-2026-91140 with Vani AIImpact, affected versions, exploitation, mitigations and detection

Coverage

oldest first · one row per article · UTC
  1. · 24d ago Canadian Centre for Cyber SecurityFirst report

    Progress security advisory (AV26-1005)

    Canada's Cyber Centre warns of CVE-2026-91140 in Progress ARCGenAI-Generator before version 2.1.

  2. · 23d ago Cyber Security News

    Critical Progress DataDirect GenAI Flaw Lets Malicious OpenAPI Files Execute OS Commands

    Progress warns CVE-2026-91140 lets malicious OpenAPI files run OS commands in DataDirect GenAI agents.

  3. · 23d ago GBHackers

    Critical Progress DataDirect GenAI Flaw Lets Attackers Execute Arbitrary OS Commands

    Progress warns CVE-2026-91140 lets crafted OpenAPI files run OS commands in DataDirect GenAI agents.

Vulnerabilities in this story

All
CVEVulnerabilityCVSSEPSSSignalsPublished
CVE-2026-91140
OS Command Injection via Malicious OpenAPI Docs in Progress ARCGenAI-Generator 2.0
9.6—