CVE-2026-91987
MediumCVSS 6.5Summary
atomic-agents-stack before 1.1.0 contains a cost-guardrail bypass in the _estimate_batch_cost function that returns zero cost for unknown models not in the pricing table. Attackers can configure deployments with unknown model identifiers to bypass daily cost caps and exceed budget limits in parallel batch operations.
Risk Assessment
An attacker can bypass daily cost caps and exceed budget limits, leading to uncontrolled spending. The risk applies to deployments using atomic-agents-stack with unknown model identifiers.
Recommendation
Update atomic-agents-stack to version 1.1.0 or later. Ensure model identifiers are present in the pricing table or that unknown models are handled appropriately.
Other vulnerabilities in atomic-agents-stack
See all- CVE-2026-91989High
atomic-agents-stack before 1.1.0 contains a path traversal vulnerability in the dashboard HTTP server that allows remote attackers to read arbitrary files by supplying directory traversal sequences in request paths. Attackers can bypass path containment checks by including '../' segments in requests to the DashboardHandler.do_GET endpoint to access files outside the intended agents_root directory.
- CVE-2026-91988High
atomic-agents-stack before 1.1.0 accepts cleartext HTTP schemes in the HTTP MCP server-registry backend factory, allowing network man-in-the-middle attackers to rewrite catalog responses. Attackers can inject arbitrary command and argument values that are spawned as local subprocesses by MCPClientPool to achieve code execution on the agent host.
Original NVD description (English source)
atomic-agents-stack before 1.1.0 contains a cost-guardrail bypass in the _estimate_batch_cost function that returns zero cost for unknown models not in the pricing table. Attackers can configure deployments with unknown model identifiers to bypass daily cost caps and exceed budget limits in parallel batch operations.