Back to overview

CVE-2026-44688

HIGH
8.4
CVSS 4.0
Description
In Eclipse Theia versions prior to 1.71.0, the AI chat agent processed workspace file and directory names as part of its prompt context without distinguishing them from system instructions. An attacker could craft a malicious repository with adversarial directory or file names that, when analyzed by the AI agent, would cause the agent to follow attacker-controlled instructions (indirect prompt injection). Combined with other AI chat features available in untrusted workspaces, this enabled attack chains leading to data exfiltration via Markdown image rendering or arbitrary command execution via task definitions.

Metadata

CVE ID
CVE-2026-44688
State
PUBLISHED
Assigner
eclipse
Reserved
2026-05-22 07:47 UTC
Published
2026-06-18 14:22 UTC
Last updated
2026-06-19 03:55 UTC
Primary CWE
CWE-1427
CWE-1427 Improper neutralization of input used for LLM promp…
Vendor / Product
Eclipse Foundation / Eclipse Theia
Sources
cve.org  ·  NVD

Severity & Metrics

8.4 HIGH CVSS 4.0
CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:A/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N
SSVC — CISA Coordinator
Exploitation
none
Automatable
no
Tech. Impact
total
Affected products (1)
VendorProductPlatformVersions
Eclipse Foundation Eclipse Theia — 0 < 1.71.0
Weakness (CWE)
CWESourceDescription
CWE-1427 cna CWE-1427 Improper neutralization of input used for LLM prompting
CWE-829 cna CWE-829 Inclusion of functionality from untrusted control sphere
CVSS scores (1)
ScoreSeverityVersionSourceVector
8.4 HIGH 4.0 cna CVSS:4.0/AV:L/AC:L/AT:N/PR:N/UI:A/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N
References (1)
Back to overview