Skyvern before 1.0.45 Sandbox Escape via TextPromptBlock (CVE-2026-82447) | HOL Guard CVE

Unknown severity · Source severity not reportedCVE-2026-82447

Skyvern before 1.0.45 Sandbox Escape via TextPromptBlock

Answer in brief

CVE-2026-82447 records a Unknown severity vulnerability in Skyvern before 1.0.45 Sandbox Escape via TextPromptBlock. The current sources do not mark it as known exploited. The current feed maps Skyvern-AI/skyvern (generic). Check affected ranges and fixed versions before updating.

Analysis pending evidence review

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

Published Aug 29, 2026Updated Aug 29, 2026Source checked Aug 29, 2026First seen by HOL Aug 29, 2026Material review Aug 29, 2026

Upstream Advisory

Key facts

Risk: Unknown severitySource severity not reported
Exploitation: Not marked as known exploited
Affected software: 1 mapped package or product
Fix availability: Available

Why this deserves its current priority

A CVSS score is not reported in the current record. 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 Skyvern-AI/skyvern (generic). Check affected ranges and fixed versions before updating.

Mapped affected packages and fixed versions
Package	Affected range	Fixed version
Skyvern-AI/skyverngeneric	>=0.2.1 <1.0.45	1.0.45

Recommended response

Check inventory. Check lockfiles and deployed manifests for Skyvern-AI/skyvern.
Review the reported fix. Update Skyvern-AI/skyvern to 1.0.45 if you use the affected versions. Test the change in a non-production environment first.

Evidence timeline and material changes

Published upstream
Aug 29, 2026
Evidence: source:cvelist:source_dates:source-dates:record
Source modified
Aug 29, 2026
Evidence: source:cvelist:source_dates:source-dates:record
First seen by HOL
Aug 29, 2026

Sources and claim methodology

Upstream source description

Skyvern before 1.0.45 contains a sandbox escape vulnerability in TextPromptBlock that renders prompts twice, first through a sandboxed Jinja environment and then through an unsandboxed environment. Attackers can inject malicious Jinja template syntax through workflow parameters or upstream block output to execute arbitrary code with server process privileges.

Quoted source text, attributed separately from HOL analysis.