CVE-2026-84887
Received Received - Intake

Denial of Service in simular-ai Agent-S via Model-generated GUI Action Execution Workflow

Publication date: 2026-09-03

Last updated on: 2026-09-03

Assigner: VulDB

Description
A vulnerability was identified in simular-ai Agent-S up to 0.3.2. Affected by this issue is some unknown functionality of the file grounding.py of the component Model-generated GUI Action Execution Workflow. The manipulation leads to denial of service. Remote exploitation of the attack is possible. The exploit is publicly available and might be used. The vendor was contacted early about this disclosure but did not respond in any way.

CVSS Scores

EPSS Scores

Probability:
Percentile:

Meta Information

Published
2026-09-03
Last Modified
2026-09-03
Generated
2026-09-03
AI Q&A
2026-09-03
EPSS Evaluated
N/A
NVD
CVE-2026-84887
EUVD
EUVD-2026-70285

Affected Vendors & Products

Showing 1 associated CPE
Vendor Product Version / Range
simular-ai agent-s to 0.3.2 (inc)

Helpful Resources

Exploitability

CWE
CWE Icon
KEV
KEV Icon
CWE ID Description
CWE-404 The product does not release or incorrectly releases a resource before it is made available for re-use.

Attack-Flow Graph

AI Quick Actions

Instant insights powered by AI
Executive Summary

This vulnerability is a denial-of-service issue in simular-ai Agent-S up to version 0.3.2. It occurs in the grounding.py file where model-generated action parameters are not properly validated before being executed via eval(). This allows crafted inputs to monopolize system resources, such as causing the agent to sleep for extended periods with commands like agent.wait(3600).

Detection Guidance

Monitor for unusually long-running processes or blocked tasks in Agent-S up to 0.3.2. Check logs for excessive resource usage from model-generated actions like agent.wait() or eval() calls. Look for pending tasks that never complete due to malicious prompts causing unbounded waits or loops.

Impact Analysis

This vulnerability can impact you by causing system slowdowns or crashes due to resource exhaustion. Attackers could send malicious prompts to block legitimate tasks, making the agent unavailable for normal operations. Systems relying on Agent-S for automation may experience degraded performance or complete failure.

Compliance Impact

This vulnerability primarily causes denial-of-service by allowing unbounded resource consumption through model-generated actions like agent.wait(3600). For compliance with GDPR, this could impact availability of services processing personal data, potentially violating Article 32 requirements for security and integrity. Under HIPAA, it may disrupt access to protected health information systems, affecting the Availability standard (45 CFR § 164.312(a)(2)(ii)). The lack of input validation and resource limits could lead to unauthorized resource exhaustion, affecting integrity and availability controls required by both standards.

Mitigation Strategies

Upgrade Agent-S to a patched version if available. Implement strict input validation for model-generated actions, enforce parameter limits (e.g., max wait time), and add per-action timeouts. Disable eval() usage for untrusted inputs and set runtime budgets to prevent resource exhaustion.

Chat Assistant

Ask questions about this CVE
Hi! I’m here to help you understand CVE-2026-84887. Ask me anything about the vulnerability, its impact, or mitigation strategies.
0/70

EPSS Chart