HighVulnerabilityLLM-specific

CVE-2025-58370: Roo Code command parsing flaw allows arbitrary command execution via prompts

Source: NVD(opens in a new tab)
Identifier: CVE-2025-58370
Published: September 5, 2025
Record updated: October 9, 2026

Known exploitation: Not listed in the CISA Known Exploited Vulnerabilities catalog at the last check.
EPSS: 0.4%

Summary

CVE-2025-58370 affects Roo Code, an AI-powered autonomous coding agent, in versions below 3.26.0. The command parsing logic mishandles Bash parameter expansion and indirect references, so when the agent is configured to auto-approve certain commands, an attacker who can influence prompts can get additional arbitrary commands executed alongside the intended one. The weakness is classified as CWE-78, OS Command Injection.

Mitigation

This is fixed in version 3.26.0.

Details

Scores and exploitation

CVSS score: 8.1(high)
EPSS: 0.4%Estimated probability that this CVE is exploited in the next 30 days, from the FIRST Exploit Prediction Scoring System. Read on November 4, 2026.

Weaknesses and techniques

Weakness (CWE): CWE-78
Attack pattern (CAPEC): CAPEC-88

Classification

Attack type: Other
Attack sophistication: Moderate
AI component targeted: Agent
Impact: Confidentiality,
Integrity
Labels: security

How this record was produced

Summaries, mitigations and labels on this page are generated automatically from the source. The linked source is authoritative. The source gives this advisory a description and no title, so the title above is a headline generated from the source text.

Original source: https://nvd.nist.gov/vuln/detail/CVE-2025-58370

First tracked: February 15, 2026 at 08:53 PM ET, from NVD/CVE Database

Classifier: claude-haiku-5-5, prompt v4

Classifier confidence: 93% (self-assessed)

Topics

AI agents

Summary

Mitigation

Topics

Related items