CVE-2026-107717
When an application renders untrusted data and passes the returned ChatMessage objects to an LLM provider, attacker-controlled JSON can cross the prompt boundary and become a system, assistant, or tool message because ChatMessage.role accepts arbitrary…
Does this matter?
Lower severity and a low EPSS score (0.34%). Track it; it rarely justifies an emergency change on its own.
Description
Banks generates meaningful LLM prompts using a simple template language. Prior to 2.5.0, Banks Prompt.chat_messages() attempts to parse every line of rendered template output as ChatMessage JSON. When an application renders untrusted data and passes the returned ChatMessage objects to an LLM provider, attacker-controlled JSON can cross the prompt boundary and become a system, assistant, or tool message because ChatMessage.role accepts arbitrary strings. This can override application instructions, alter the intended prompt structure, or confuse downstream tool and message handling. This issue is fixed in version 2.5.0.
- CVSS 3.1
- 6.5 MEDIUMCVSS:3.1/AV:N/AC:L/PR:N/UI:N/S:U/C:L/I:L/A:N
- EPSS
- 0.34% probability · 27th percentile
- Public exploits
- None in Exploit-DB
- CISA KEV
- Not listed
- Weakness
- CWE-20
- Source
- security-advisories@github.com
EPSS moved from 0.3% on 9 October 2026 to 0.3% on 26 October 2026.
References
- https://github.com/masci/banks/commit/02172b816fb84f6a824cc09a8aca7416f53c12cb
- https://github.com/masci/banks/pull/78
- https://github.com/masci/banks/releases/tag/v2.5.0
- https://github.com/masci/banks/security/advisories/GHSA-hmq2-7hp6-7crh
- https://github.com/masci/banks/security/advisories/GHSA-hmq2-7hp6-7crh
Source: NVD record, EPSS from FIRST.org, KEV from CISA, exploits from Exploit-DB. Refreshed daily. Download this record as JSON.