Medium · CVSS 6.5CVE-2026-107717GHSA-HMQ2-7HP6-7CRH

Banks: User-controlled prompt input can be parsed as privileged chat messagesCVE-2026-107717

Answer in brief

CVE-2026-107717 records a Medium severity (CVSS 6.5) vulnerability in Banks: User-controlled prompt input can be parsed as privileged chat messages. The current sources do not mark it as known exploited. The current feed maps masci/banks (generic), banks (pip). Check affected ranges and fixed versions before updating.

Analysis pending evidence review

HOL Guard separates source facts from reviewed analysis. See the methodology.

Published Oct 8, 2026Updated Oct 8, 2026Source checked Oct 8, 2026First seen by HOL Oct 8, 2026Material review Oct 8, 2026

Upstream Advisory

Key facts

Risk: Medium · CVSS 6.5
Exploitation: Not marked as known exploited
Affected software: 2 mapped packages or products
Fix availability: Available

Why this deserves its current priority

CVSS is 6.5. The current sources do not mark it as known exploited. Treat this as a source-backed prioritization signal, not a statement about your environment.

Analysis status

Analysis pending evidence review

Factual feed record only; HOL analysis is not approved for indexing. Read the methodology.

Affected scope and exposure questions

The current feed maps masci/banks (generic), banks (pip). Check affected ranges and fixed versions before updating.

Mapped affected packages and fixed versions
Package	Affected range	Fixed version
masci/banksgeneric	<2.5.0	2.5.0
bankspip	<=2.4.5	2.5.0

Recommended response

Check inventory. Check lockfiles and deployed manifests for masci/banks, banks.
Review the reported fix. Update masci/banks to 2.5.0; banks to 2.5.0 if you use the affected versions. Test the change in a non-production environment first.

Evidence timeline and material changes

Published upstream
Oct 8, 2026
Evidence: source:cvelist:source_dates:source-dates:record
Source modified
Oct 8, 2026
Evidence: source:cvelist:source_dates:source-dates:record
First seen by HOL
Oct 8, 2026

Sources and claim methodology

Upstream source 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.

Quoted source text, attributed separately from HOL analysis.