CVE-2026-82580 AshAi echoes raw tool exception messages into the conversation, disclosing internal details
Attackers can trigger tool exceptions to leak sensitive internal error details, such as database constraints or query fragments, to chat users.
- Exploited in the wild
- Not known
- Login needed
- Any signed-in user
- Attackable
- Over the internet
- Victim action needed
- Not needed
- Fix
- Available
What is CVE-2026-82580?
CVE-2026-82580 is a medium-severity vulnerability in ash-project ash_ai, published on 31 Aug 2026. Any signed-in user can exploit it over the network, with no action from the victim.
Official description
Generation of Error Message Containing Sensitive Information vulnerability in ash-project ash_ai discloses internal error text to chat users.
In AshAi.ToolLoop and AshAi.Tools, an exception raised while executing a tool was serialized verbatim with Exception.message/1 into the tool-result content. That content is appended to the conversation, emitted as a {:tool_result, ...} stream event, and sent back to the model, which typically relays it to the user. No filtering happened first, so anything raised inside a tool callback or lifecycle hook (database constraint messages, adapter errors, query fragments, policy or validation internals) was echoed as-is. A chat user who can steer tool arguments into a raising code path receives the raw internal text. The fix routes raised tool errors through the same safe formatter used for other tool errors.
This issue affects ash_ai: from 0.6.0 before 1.0.0.
Weakness: CWE-209
Which versions are affected by CVE-2026-82580?
1 open source package ship an affected version, including ash_ai.
| Package | Affected | Fixed in |
|---|---|---|
ash_ai Hex | From 0.6.0 | 1.0.0 |
How do I fix CVE-2026-82580?
Upgrade ash_ai to 1.0.0 or later.
Is CVE-2026-82580 being exploited?
It is not confirmed as exploited in the wild, but public exploit code for CVE-2026-82580 exists.
Sources
Also known as EEF-CVE-2026-82580, GHSA-5747-6mpw-6m9c.