llamaCVEVulnerability Intelligence
/
copilot-cli

CVE-2026-29783

GitHub Copilot CLI allows for dangerous shell expansion patterns that enable arbitrary command execution

Specific setups onlyParche disponibleOS command injection
Descargar PNG JSON
CVSS 4.0
7.5
Alta
CVSS:4.0/AV:N/AC:L/AT:P/PR:N/UI:A/VC:H/VI:H/VA:H/SC:N/SI:N/SA:N Fuente: CNA
ExplotaciónSin evidenciaSSVC: none · no figura en CISA KEV
Exploits públicos0Ninguno conocido
Parche1 versiónPrimer parche el mismo día de la divulgación
Exposición257 dDesde la divulgación, 6 mar 2026, 11:39 a. m.
Detección241 dDesde la divulgación hasta su ingreso al índice
Decisión SSVCSeguir de cerca
Monitorear y programar la remediación
Sin explotación conocida, no automatizable e impacto técnico total. Hay versiones corregidas disponibles.
Timeline

Ciclo de vida de la vulnerabilidad

Hoy
Divulgación
Publicada
Evaluación SSVC
Modificada
Explotación
Remediación
Parche 0.0.423
Monitoreo
Detectada
Verificada
6 mar 26
11 mar 26
3 nov 26
16 nov 26
19 nov 26
6 mar 26
10 may 26
13 jul 26
16 sept 26
19 nov 26
FechaFaseEventoDetalle
Divulgación Publicada CVE-2026-29783 divulgada · CVSS 7.5 (CNA)
Remediación Parche 0.0.423 @github/copilot · 0 d desde divulgación
Divulgación Evaluación SSVC CISA-ADP · explotación ninguna
Monitoreo Detectada Ingresó al índice
Divulgación Modificada Última actualización del registro
Monitoreo Verificada Última verificación contra las fuentes
Resumen

The shell tool within GitHub Copilot CLI versions prior to and including 0.0.422 can allow arbitrary code execution through crafted bash parameter expansion patterns. An attacker who can influence the commands executed by the agent (e.g., via prompt injection through repository files, MCP server responses, or user instructions) can exploit bash parameter transformation operators to execute hidden commands, bypassing the safety assessment that classifies commands as "read-only." This has been patched in version 0.0.423. The vulnerability stems from how the CLI's shell safety assessment evaluates commands before execution. The safety layer parses and classifies shell commands as either read-only (safe) or write-capable (requires user approval). However, several bash parameter expansion features can embed executable code within arguments to otherwise read-only commands, causing them to appear safe while actually performing arbitrary operations. The specific dangerous patterns are ${var@P}, ${var=value} / ${var:=value}, ${!var}, and nested $(cmd) or <(cmd) inside ${...} expansions. An attacker who can influence command text sent to the shell tool - for example, through prompt injection via malicious repository content (README files, code comments, issue bodies), compromised or malicious MCP server responses, or crafted user instructions containing obfuscated commands - could achieve arbitrary code execution on the user's workstation. This is possible even in permission modes that require user approval for write operations, since the commands can appear to use only read-only utilities to ultimately trigger write operations. Successful exploitation could lead to data exfiltration, file modification, or further system compromise.

¿Qué significa?Copilot CLI can be tricked into running shell commands that an attacker chose, because certain shell expansion patterns are not handled safely. This can bypass the tool's read-only safety classification and its approval requirements, leading to arbitrary code execution on the user's workstation.
¿Quién está afectado?The affected software is GitHub Copilot CLI, the @github/copilot npm package, in versions up to and including 0.0.422. The affected case requires that an attacker can influence the shell commands the agent runs, for example through prompt injection in repository files, MCP server responses, or user instructions. To check the installed version, look at the version reported by the Copilot CLI or the installed @github/copilot npm package.
¿Qué hacer?Upgrade GitHub Copilot CLI to version 0.0.423 or later. No workaround is listed in the facts.
Contexto

Condiciones y alcance

Condiciones necesarias para que el fallo sea explotable
01
Affected software is GitHub Copilot CLI (@github/copilot npm package) versions prior to and including 0.0.422.“Affected versions <= 0.0.422”
02
Attacker must be able to influence the shell commands the agent executes, e.g. via prompt injection through repository files, MCP server responses, or user instructions.“An attacker who can influence the commands executed by the agent (e.g., via prompt injection through repository files, MCP server responses, or user instructions)”
Vector CVSS 4.0

Desglose de métricas

Opción resaltada = valor del vector · izquierda = peor caso · CNA
Vector de ataqueAV:N
RedAdyacenteLocalFísica
ComplejidadAC:L
BajaAlta
Requisitos de ataqueAT:P
NingunoPresentes
PrivilegiosPR:N
NingunoBajosAltos
InteracciónUI:A
NingunaPasivaActiva
Confid. (sistema)VC:H
AltoBajoNinguno
Integridad (sistema)VI:H
AltoBajoNinguno
Disponib. (sistema)VA:H
AltoBajoNinguno
Confid. (posterior)SC:N
AltoBajoNinguno
Integridad (posterior)SI:N
AltoBajoNinguno
Disponib. (posterior)SA:N
AltoBajoNinguno
Versiones

Rangos afectados

Afectada Corregida Última
copilot-cli≤ 0.0.422
@github/copilot< 0.0.423
0.0.422
0.0.423
1.0.112
Remediación

Estado del parche por paquete

Paquete / productoEstadoCorregida enDesde divulgaciónÚltima versión
npm@github/copilot
Corregida 0.0.423 0 d 1.0.112 no afectada
Versiones corregidas0.0.423
Explotación

Arsenal público

Sin exploits, módulos ni PoC públicos conocidos a la fecha de verificación.

Referencias
Fabricantehttps://github.com/github/copilot-cli/security/advisories/GHSA-g8r9-g2v8-jv6f
Otrahttps://github.com/github/copilot-cli/releases/tag/v0.0.423

Las referencias externas se muestran para copiar: cve.llama.ec no lo redirige a otros sitios.