CVE-2026-82580

AshAi tool exceptions could leak internal error details

ash-project · ash_ai

First documented disclosure · Review checkpoint

Summary

Raised tool errors could be exposed through the model conversation and stream.

Severity and exploitation evidence

Medium · 5.3/10 · CVSS 4.0. Severity source.

Exploitation not established: Malicious exploitation is not established by the reviewed originals. Public demonstrations and catalog absence do not establish exploitation status.

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

ToolLoop and Tools serialized raw exception messages without the ordinary safe formatter.

Conditions required

A chat user can steer a tool callback or lifecycle hook into an exception path.

Potential impact

Database constraints, query fragments and policy or validation internals can reach the model and user.

Affected versions and configurations

ash_ai: 0.6.0 through versions before 1.0.0.

Fix and mitigation status

1.0.0.

Practical next steps

Apply the issue-specific correction: 1.0.0.

Operating lesson

Apply one safe error formatter to every tool failure path.

Evidence and limits

Earliest reviewed public evidence 2026-08-30; CVE publication 2026-08-31. CVSS 4.0 5.3 is attributed to EEF in the CNA record.

Sources