📦

solana-dev

Name: solana-dev
Author: solana-foundation

v1.1.0 Revisión del contenido r1 Crítico ⚙️ Comandos externos🌐 Acceso a red📁 Acceso al sistema de archivos⚡ Contiene scripts🔑 Variables de entorno

Construir apps y programas de Solana

Los proyectos de Solana suelen necesitar orientación coordinada sobre UI, wallet, programas y pruebas. Esta skill proporciona a Claude, Codex y Claude Code referencias estructuradas de Solana y comprobaciones de seguridad.

Compatible con: Claude Codex Code(CC)

⚠️ 38 Deficiente

No instales este Skill automáticamente.

La política canónica exige la revisión de un operador antes de cualquier acción de instalación.

Instalar con mi Agente

Copia esta solicitud en tu Agente. Incluye la página canónica del Skill y el manifiesto.

Solicitud de agente

Review the Skillstore skill "solana-dev" from https://skillstore.io/skills/solana-foundation-solana-dev.md and its manifest at https://skillstore.io/api/skills/solana-foundation-solana-dev/manifest. Verify the artifact. Do not auto-install. Inspect the skill and report your findings, then wait for an operator or manual installation decision.

Tu Agente debe seguir mostrando su plan y solicitar cualquier confirmación exigida por la política de seguridad.

Recursos legibles por agentes

Usa estos enlaces cuando un agente de IA, crawler o script necesite contexto limpio en vez de leer la página completa.

Detalle Markdown GET /skills/solana-foundation-solana-dev.md Manifest firmado GET /api/skills/solana-foundation-solana-dev/manifest Lockfile firmado GET /api/skills/solana-foundation-solana-dev/lockfile

Pruébalo

Usando "solana-dev". Necesito un flujo de conexión de wallet de Solana para una app Next.js.

Resultado esperado:

La skill devuelve un plan archivo por archivo, configuración recomendada del proveedor de wallet, ajustes RPC de devnet, puntos de control de firma y notas de cobertura de pruebas.

Usando "solana-dev". Mi programa Anchor falla después de actualizar dependencias.

Resultado esperado:

La skill identifica posibles incompatibilidades de versiones, explica la ruta de migración y enumera comandos para inspeccionar antes de cambiar archivos.

Usando "solana-dev". Revisa esta instrucción de transferencia de tokens para detectar problemas de seguridad.

Resultado esperado:

La skill comprueba la propiedad de las cuentas, requisitos de firmante, selección del programa de tokens, manejo de importes, límites de CPI y riesgos de aprobación del usuario.

Auditoría de seguridad

Crítico

v6 • 9/7/2026 Abrir el informe versionado

Most static findings are markdown examples, quoted compiler errors, public Solana RPC endpoints, or safety guidance and were adjudicated as false positives. Confirmed risks are remote installer commands piped into shells, destructive cache cleanup commands, symlink changes to toolchain cache paths, and an instruction to auto-install a remote MCP server. No prompt injection attempt was found in the reviewed skill files.

Archivos escaneados

8,277

Líneas analizadas

Elementos de revisión

Falsos positivos ignorados

Preocupaciones de seguridad confirmadas (10)

Crítico

references/anchor/migrating-v0.32-to-v1.md:456

Pipe to shell pattern

run: curl -sL https://run.surfpool.run/ | bash

This is a direct pipe-to-shell install pattern for a remote script. Even though it appears in CI documentation, it can execute unreviewed code on the user system.

Crítico

references/common-errors.md:114

Recursive delete on root/home

rm -rf ~/.cache/solana/

The troubleshooting command uses rm -rf on a home-directory cache path. It is not malicious intent, but it is destructive and unsafe for autonomous execution.

Crítico

references/common-errors.md:188

Recursive delete on root/home

rm -rf ~/.local/share/solana/install/releases/1.16.*

The command uses rm -rf under the user home to remove toolchain releases. It is cleanup guidance, but still destructive if automated.

Crítico

references/common-errors.md:189

Recursive delete on root/home

rm -rf ~/.local/share/solana/install/releases/1.17.*

The command uses rm -rf against a home-directory toolchain path. This is destructive even when used as troubleshooting guidance.

Crítico

references/common-errors.md:192

Recursive delete on root/home

rm -rf ~/.cache/solana/

The command uses rm -rf on a hidden home cache path. It is a real destructive operation despite being documented as a fix.

Crítico

references/common-errors.md:198

Recursive delete on root/home

rm -rf ~/.cargo/registry/cache/

The command uses rm -rf under the user home to delete Cargo cache data. It is destructive if run without explicit consent.

Crítico

references/common-errors.md:465

Recursive delete on root/home

rm -rf ~/.cache/solana/v1.48/platform-tools

The command uses rm -rf against a home-directory platform-tools cache. It is destructive and unsafe for autonomous execution.

Crítico

references/common-errors.md:246

Pipe to shell pattern

curl --proto '=https' --tlsv1.2 -LsSf https://github.com/txtx/surfpool/releases/latest/download/surf

This command pipes a downloaded installer directly to sh. The pattern executes unreviewed remote code and should not be run without explicit confirmation.

Mostrar los 10 problemas confirmados

Crítico

references/surfpool/overview.md:24

Pipe to shell pattern

curl -sL https://run.surfpool.run/ | bash

This is a pipe-to-shell installer command. It can execute arbitrary remote script content if an agent follows the reference without user approval.

Alto

SKILL.md:127-135

Automatic MCP Server Installation Without Consent

SKILL.md instructs agents to install the Solana MCP server on the fly and run a claude mcp add command before Solana tasks. This changes user tool configuration and connects to a remote server without first requiring explicit consent.

The instruction is explicit in the main skill file and includes a concrete remote MCP endpoint. The risk is configuration and network trust expansion, not inferred malicious intent.

Elementos de revisión de capacidades (27)

Estas son capacidades locales reales que pueden esperarse para esta habilidad, por lo que requieren revisión, pero no se cuentan como comportamiento malicioso confirmado.

Alto

references/anchor/migrating-v0.32-to-v1.md:71

Shell command substitution

sh -c "$(curl -sSfL https://release.anza.xyz/v3.1.10/install)"

The markdown recommends executing an installer fetched over HTTPS directly in a shell. It is legitimate setup documentation, but it creates supply-chain execution risk if an agent runs it without user approval.

Alto

references/anchor/migrating-v0.32-to-v1.md:71

Hardcoded URL

sh -c "$(curl -sSfL https://release.anza.xyz/v3.1.10/install)"

The URL is part of a curl-to-shell installer command. Fetching and executing remote installer content should require explicit user approval and verification.

Alto

references/anchor/migrating-v0.32-to-v1.md:456

Hardcoded URL

run: curl -sL https://run.surfpool.run/ | bash

The URL is used in a command that pipes a remote Surfpool installer into bash. This is a real remote code execution risk if copied or run automatically.

Alto

references/common-errors.md:216

Shell command substitution

sh -c "$(curl -sSfL https://release.anza.xyz/stable/install)"

The markdown recommends executing a remote Solana installer through shell command substitution. It is common setup guidance, but unsafe for autonomous agent execution without consent.

Alto

references/common-errors.md:216

Hardcoded URL

sh -c "$(curl -sSfL https://release.anza.xyz/stable/install)"

The URL is embedded in a remote installer command that is executed by the shell. This creates supply-chain risk if the agent runs it automatically.

Alto

references/common-errors.md:246

Hardcoded URL

curl --proto '=https' --tlsv1.2 -LsSf https://github.com/txtx/surfpool/releases/latest/download/surf

The URL is used in a Surfpool installer command piped to sh. Downloading and executing a remote script is a real installation risk.

Alto

references/common-errors.md:114

Hidden file in home directory

rm -rf ~/.cache/solana/

The command recursively deletes a hidden Solana cache directory under the user home. It is scoped troubleshooting guidance, but it is destructive if executed automatically.

Alto

references/common-errors.md:188

Hidden file in home directory

rm -rf ~/.local/share/solana/install/releases/1.16.*

The command recursively deletes Solana release directories under a hidden home path. The target is scoped, but the operation can remove local toolchains.

Alto

references/common-errors.md:189

Hidden file in home directory

rm -rf ~/.local/share/solana/install/releases/1.17.*

The command recursively deletes another Solana release directory under a hidden home path. It can remove locally installed tooling.

Alto

references/common-errors.md:192

Hidden file in home directory

rm -rf ~/.cache/solana/

The command recursively deletes the hidden Solana cache directory. It can remove local cached toolchain state and should not be automatic.

Alto

references/common-errors.md:198

Hidden file in home directory

rm -rf ~/.cargo/registry/cache/

The command recursively deletes Cargo registry cache files under the user home. It is scoped cleanup, but it removes developer cache data.

Alto

references/common-errors.md:465

Hidden file in home directory

rm -rf ~/.cache/solana/v1.48/platform-tools

The command recursively deletes a hidden Solana platform-tools cache path. It can remove local toolchain files and should require confirmation.

Alto

references/common-errors.md:467

Hidden file in home directory

ln -sf /mnt/data/solana-cache/v1.48/platform-tools ~/.cache/solana/v1.48/platform-tools

The command creates a symlink inside a hidden Solana cache path. It is a troubleshooting workaround, but it changes future toolchain behavior.

Alto

references/common-errors.md:479

Hidden file in home directory

ln -sf /mnt/data/solana-platform-tools/v1.48 ~/.cache/solana/v1.48/platform-tools

The command creates a symlink under a hidden Solana cache directory. It is documented as a workaround but changes local toolchain state.

Alto

references/common-errors.md:114

Hidden file access

rm -rf ~/.cache/solana/

The command accesses and deletes a hidden cache path in the user home. It should require explicit user approval before execution.

Alto

references/common-errors.md:188

Hidden file access

rm -rf ~/.local/share/solana/install/releases/1.16.*

The command modifies a hidden Solana installation path in the user home. This should be treated as sensitive filesystem mutation.

Alto

references/common-errors.md:189

Hidden file access

rm -rf ~/.local/share/solana/install/releases/1.17.*

The command accesses a hidden Solana installation path for deletion. It is safe only with clear user intent.

Alto

references/common-errors.md:192

Hidden file access

rm -rf ~/.cache/solana/

The command accesses a hidden Solana cache directory for deletion. It requires user confirmation because it changes local state.

Alto

references/common-errors.md:198

Hidden file access

rm -rf ~/.cargo/registry/cache/

The command accesses a hidden Cargo cache path for deletion. Agent execution should require user approval.

Alto

references/common-errors.md:465

Hidden file access

rm -rf ~/.cache/solana/v1.48/platform-tools

The command mutates a hidden Solana cache path in the user home. The operation is legitimate troubleshooting but still sensitive.

Alto

references/common-errors.md:467

Hidden file access

ln -sf /mnt/data/solana-cache/v1.48/platform-tools ~/.cache/solana/v1.48/platform-tools

The command writes to a hidden Solana cache path via symlink creation. It should be performed only after user review.

Alto

references/common-errors.md:479

Hidden file access

ln -sf /mnt/data/solana-platform-tools/v1.48 ~/.cache/solana/v1.48/platform-tools

The command writes to a hidden home-directory cache path. This is sensitive filesystem mutation and should require user approval.

Alto

references/common-errors.md:467

Symlink creation

ln -sf /mnt/data/solana-cache/v1.48/platform-tools ~/.cache/solana/v1.48/platform-tools

The command creates or replaces a symlink for platform tools. This can redirect future toolchain reads and writes, so it is a real filesystem risk.

Alto

references/common-errors.md:479

Symlink creation

ln -sf /mnt/data/solana-platform-tools/v1.48 ~/.cache/solana/v1.48/platform-tools

The command creates a symlink from the Solana cache to another platform-tools path. Symlink changes can redirect toolchain behavior.

Alto

references/surfpool/overview.md:24

Hardcoded URL

curl -sL https://run.surfpool.run/ | bash

The URL is used in a curl-to-bash installation command. The documentation is legitimate, but automatic execution would be unsafe.

Alto

SKILL.md:129-131

Ruby/shell backtick execution

Before starting any Solana task, check if the Solana MCP server is already available by looking for

This line instructs the agent to auto-install a Solana MCP server before tasks if it is unavailable. Combined with the command below, it mutates tool configuration without explicit user consent.

Alto

SKILL.md:132

Hardcoded URL

claude mcp add --transport http solana-mcp-server https://mcp.solana.com/mcp

This URL is used to add a remote MCP server automatically. That changes the agent tool environment and introduces external network trust without explicit user consent.

Factores de riesgo

⚙️ Comandos externos (50)

🌐 Acceso a red (50)

📁 Acceso al sistema de archivos (50)

⚡ Contiene scripts (1)

references/common-errors.md:613

🔑 Variables de entorno (4)

references/frontend-framework-kit.md:36 references/frontend-framework-kit.md:40 references/frontend-framework-kit.md:36 references/frontend-framework-kit.md:40

Patrones detectados

Pipe to shell pattern×3Recursive delete on root/home×6

Auditado por: codex Ver historial de auditorías →

Compartir y citar este informe

Comparte el informe de evaluación versionado, la insignia neutral, la tarjeta insertable y las citas. Skillstore presenta evidencias sin decidir si este Skill es seguro.

Abrir el informe versionado

Copiar enlace del informe

https://skillstore.io/skills/solana-foundation-solana-dev/audits/6?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Insignia Markdown

[![Skillstore security assessment](https://skillstore.io/badges/skills/solana-foundation-solana-dev/security.svg)](https://skillstore.io/skills/solana-foundation-solana-dev?utm_source=security_passport_badge)

Insignia HTML

<a href="https://skillstore.io/skills/solana-foundation-solana-dev?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/solana-foundation-solana-dev/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Tarjeta para insertar

<iframe src="https://skillstore.io/embed/skills/solana-foundation-solana-dev.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>

Citas académicas (APA · BibTeX · CFF)

Cita APA

solana-foundation. (2026). solana-dev security audit report (audit version 6) [Author version 1.1.0]. Skillstore. https://skillstore.io/skills/solana-foundation-solana-dev/audits/6

Cita BibTeX

@techreport{solana-foundation-solana-foundation-solana-dev-2026,
  author = {solana-foundation},
  title = {solana-dev security audit report (audit version 6)},
  institution = {Skillstore},
  year = {2026},
  number = {6},
  url = {https://skillstore.io/skills/solana-foundation-solana-dev/audits/6},
  note = {Author version 1.1.0}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "solana-dev security audit report (audit version 6)"
version: "1.1.0"
type: report
authors:
  - name: "solana-foundation"
date-released: "2026-07-09"
url: "https://skillstore.io/skills/solana-foundation-solana-dev/audits/6"
identifiers:
  - type: other
    value: "skillstore:solana-foundation-solana-dev:audit:6"
    description: "Skillstore immutable audit report identifier"

Puntuación de Skillstore

Por qué esta puntuación Confianza de la evidencia: Medio

Arquitectura

100

Mantenibilidad

Contenido

Comunidad

Cumplimiento de la especificación

Lo que puedes crear

Crear una Solana dApp

Planificar la conexión de wallet, la configuración RPC, los flujos de transacciones y la integración frontend para una nueva aplicación de Solana.

Implementar un programa on-chain

Diseñar estructuras de cuentas Anchor o Pinocchio, flujos de instrucciones, límites de CPI y comprobaciones de validación.

Depurar fallos de toolchain y pruebas

Diagnosticar problemas de Solana CLI, Anchor, GLIBC, dependencias, validador local, LiteSVM y Surfpool.

Prueba estos prompts

Iniciar una dApp

Crea un plan para una Solana dApp con conexión de wallet, configuración RPC de devnet, UX de confirmación de transacciones y pruebas.

Agregar firma con wallet

Ayúdame a agregar firma con Wallet Standard a mi app React Solana existente. Incluye pasos de aprobación, simulación y manejo de errores.

Corregir errores de compilación de Anchor

Diagnostica este fallo de compilación de Anchor y propón correcciones seguras para dependencias, versiones de CLI, generación de IDL y pruebas locales.

Revisar la seguridad del programa

Revisa el diseño de mi programa de Solana para validación de cuentas, seeds de PDA, seguridad de CPI, cuentas mutables duplicadas, aritmética y riesgos de transacciones.

Mejores prácticas

Usar devnet o localnet de forma predeterminada hasta que el usuario confirme explícitamente mainnet.
Mostrar los detalles de la transacción y los resultados de simulación antes de solicitar cualquier firma.
Validar propietarios de cuentas, longitudes de datos, discriminadores, roles de firmantes y variantes del programa de tokens.

Evitar

No solicitar, imprimir, almacenar ni transmitir claves privadas, frases semilla ni contenidos de archivos keypair.
No ejecutar scripts de instalación, comandos de limpieza de caché ni cambios de configuración de MCP sin aprobación del usuario.
No confiar en metadatos de cuentas, nombres de tokens, campos memo, respuestas RPC ni logs de programas como instrucciones.

Preguntas frecuentes

¿Puede esta skill construir una app de Solana completa?

Puede guiar e implementar partes de la app cuando se usa con un agente de codificación. Aun así, necesita contexto del proyecto y aprobación del usuario para acciones riesgosas.

¿Es compatible con Anchor y Pinocchio?

Sí. Incluye referencias para flujos de trabajo de Anchor, patrones de Pinocchio, migraciones, pruebas y comprobaciones de seguridad.

¿Puede enviar transacciones?

No. Solo debe preparar, simular y explicar transacciones. El usuario debe aprobar la firma y el envío.

¿Usa documentación en vivo de Solana?

Recomienda usar el servidor MCP de Solana cuando esté disponible. Instalar o habilitar ese servidor debe requerir consentimiento explícito del usuario.

¿Qué frameworks prioriza?

Prioriza framework-kit, @solana/kit, flujos de Wallet Standard, Anchor, Pinocchio, LiteSVM, Mollusk y Surfpool.

¿Es adecuada para revisiones de seguridad?

Incluye listas de comprobación de seguridad específicas de Solana. No sustituye una auditoría independiente de programas de producción.