스킬 gasp-diagnostics

📦

gasp-diagnostics

Name: gasp-diagnostics
Author: AcceleratedIndustries

콘텐츠 리비전 r1 중간 위험 🌐 네트워크 접근📁 파일 시스템 접근⚙️ 외부 명령어

Diagnostica hosts Linux con métricas GASP

Los problemas de rendimiento en Linux son difíciles de evaluar a partir de datos de monitorización sin procesar. Esta skill ayuda a Claude, Codex y Claude Code a convertir métricas GASP en hallazgos diagnósticos claros.

지원: Claude Codex Code(CC)

⚠️ 50 나쁨

Maintain this Skill? Claim it ⚠️ 50 나쁨

내 에이전트로 설치

이 요청을 에이전트에 복사하세요. 공식 Skill 페이지와 매니페스트가 포함되어 있습니다.

에이전트 요청

Review the Skillstore skill "gasp-diagnostics" from https://skillstore.io/skills/acceleratedindustries-gasp-diagnostics.md and its manifest at https://skillstore.io/api/skills/acceleratedindustries-gasp-diagnostics/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

에이전트는 계속해서 계획을 표시하고 보안 정책에서 요구하는 확인을 요청해야 합니다.

에이전트가 읽기 쉬운 리소스

AI 에이전트, 크롤러, 스크립트가 전체 페이지 대신 정리된 컨텍스트만 필요할 때 이 링크를 사용하세요.

Markdown 상세 GET /skills/acceleratedindustries-gasp-diagnostics.md 서명된 manifest GET /api/skills/acceleratedindustries-gasp-diagnostics/manifest 서명된 lockfile GET /api/skills/acceleratedindustries-gasp-diagnostics/lockfile

테스트해 보기

"gasp-diagnostics" 사용 중입니다. Comprueba hyperion por mí.

예상 결과:

Problema encontrado: hyperion está saturado de CPU en relación con su número de núcleos. El principal contribuyente es un proceso de compilación, así que espera o reduce los trabajos en paralelo.

"gasp-diagnostics" 사용 중입니다. Compara proxmox1 y proxmox2.

예상 결과:

proxmox1 coincide con su línea base esperada de virtualización.
proxmox2 es el valor atípico, con alta presión de memoria y actividad reciente de migración.
Siguiente paso: comprueba la asignación de memoria y las migraciones recientes en proxmox2.

"gasp-diagnostics" 사용 중입니다. Aquí está la salida GASP pegada de mi host de base de datos.

예상 결과:

El host de base de datos muestra un crecimiento de memoria por encima de la línea base con actividad de swap. Revisa las consultas de larga duración y reinicia solo después de confirmar que es seguro.

보안 감사

중간 위험

v9 • 2026. 7. 7. 버전별 보고서 열기

Most external-command and hidden-file alerts are false positives from Markdown formatting or documented Claude skill installation. The confirmed risks are intentional HTTP fetches of GASP metrics from hostnames or IP addresses, which can expose internal telemetry and support broad host diagnostics. No prompt injection text was found in the reviewed files.

스캔된 파일

1,126

분석된 줄 수

검토 항목

오탐 무시됨

확인된 보안 우려 사항 (8)

중간

SKILL.md:16

System reconnaissance

web_fetch("http://{hostname}:8080/metrics")

The instruction performs a direct metrics fetch from a supplied host. This can reveal system health, processes, services, and network state to the assistant.

중간

SKILL.md:12-17 SKILL.md:76-126 README.md:75-83

Sensitive Host Telemetry Exposure

The skill directs agents to fetch GASP metrics over HTTP and analyze process, network, service, log, GPU, and active window fields. These fields can expose host names, workloads, service state, and desktop activity to the assistant session.

The fetch instruction and metric interpretation sections explicitly cover host telemetry fields. The risk is data exposure rather than the mere presence of a URL.

중간

SKILL.md:32 SKILL.md:43-53 README.md:121-125

Broad Internal Host Targeting

The trigger examples include local hostnames, IP addresses, localhost behavior, and multi-host comparisons. This can turn broad requests into internal host enumeration unless the agent asks for clear user authorization.

The examples directly instruct active fetching for IPs and multiple hosts. Legitimate administrative use is likely, but authorization boundaries are not stated.

낮음

README.md:80

System reconnaissance

web_fetch("http://hostname:8080/metrics")

The README shows active retrieval of metrics from a named host. That is legitimate for this diagnostic skill, but it is still system reconnaissance capability.

낮음

README.md:125

System reconnaissance

- Hosts accessible via **hostname, IP, or .local mDNS**

The requirements explicitly support targets by hostname, IP address, or .local mDNS name. This documents internal host targeting, which is a reconnaissance-relevant capability.

낮음

scripts/fetch-gasp.sh:6

System reconnaissance

# fetch-gasp.sh <hostname> [port]

The usage line is part of a script whose purpose is fetching GASP metrics for a supplied hostname. It supports legitimate diagnostics but exposes host reconnaissance behavior.

낮음

scripts/fetch-gasp.sh:24

System reconnaissance

echo "Usage: $0 <hostname> [port]"

The usage output prompts for a hostname that the script later uses for a metrics request. The risk is low, but it confirms host-targeted diagnostic behavior.

낮음

SKILL.md:3

System reconnaissance

description: System diagnostics using GASP (General AI Specialized Process monitor). Use when user a

The skill description says it can actively fetch GASP metrics and diagnose hosts. That is intended behavior, but it is still system reconnaissance capability.

기능 검토 항목 (6)

이 스킬에 대해 예상될 수 있는 실제 로컬 기능이므로 검토가 필요하지만, 확인된 악의적 동작으로는 집계되지 않습니다.

중간

scripts/fetch-gasp.sh:34

Hardcoded URL

URL="http://${HOSTNAME}:${PORT}/metrics"

The script constructs an HTTP metrics URL from user-supplied host and port values and then fetches it. This enables arbitrary network egress to host telemetry endpoints.

중간

SKILL.md:16

Hardcoded URL

web_fetch("http://{hostname}:8080/metrics")

The skill instructs the agent to call web_fetch against http://{hostname}:8080/metrics. This is active network access to potentially sensitive system metrics.

중간

SKILL.md:22

Hardcoded IP address

- IP addresses: `192.168.1.100`

The supported host formats include direct IP addresses. In context, this allows the skill to target internal IP-based hosts for metrics collection.

중간

SKILL.md:52

Hardcoded IP address

- "Diagnose 192.168.1.50"

The trigger examples include diagnosing a private IP address. This confirms the skill may turn user prompts into internal metrics fetches.

낮음

README.md:80

Hardcoded URL

web_fetch("http://hostname:8080/metrics")

README documents a web_fetch call to a GASP metrics URL on a supplied host. This is intentional network access to system telemetry, so it should remain visible as a network risk.

낮음

README.md:134

Hardcoded URL

Claude: [Fetches http://accelerated.local:8080/metrics]

The example fetches a metrics endpoint on an internal .local host over HTTP. This can disclose host telemetry to the assistant session even when used for legitimate diagnostics.

위험 요인

🌐 네트워크 접근 (8)

README.md:80 README.md:134 scripts/fetch-gasp.sh:34 scripts/fetch-gasp.sh:12 SKILL.md:225 SKILL.md:16 SKILL.md:22 SKILL.md:52

📁 파일 시스템 접근 (11)

README.md:24 README.md:27 README.md:30 README.md:24 README.md:27 README.md:30 README.md:24 README.md:27 README.md:30 scripts/fetch-gasp.sh:40 scripts/fetch-gasp.sh:42

⚙️ 외부 명령어 (39)

감지된 패턴

System reconnaissance×6

감사자: codex 감사 이력 보기 →

이 보고서 공유 및 인용

버전이 지정된 평가 보고서, 중립 배지, 임베드 카드 및 인용을 공유하세요. Skillstore는 근거를 보고하지만 이 Skill의 안전 여부를 판단하지 않습니다.

버전별 보고서 열기

보고서 링크 복사

https://skillstore.io/skills/acceleratedindustries-gasp-diagnostics/audits/9?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown 배지

[![Skillstore security assessment](https://skillstore.io/badges/skills/acceleratedindustries-gasp-diagnostics/security.svg)](https://skillstore.io/skills/acceleratedindustries-gasp-diagnostics?utm_source=security_passport_badge)

HTML 배지

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

임베드 카드

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

학술 인용 (APA · BibTeX · CFF)

APA 인용

AcceleratedIndustries. (2026). gasp-diagnostics security audit report (audit version 9) [Author version unspecified]. Skillstore. https://skillstore.io/skills/acceleratedindustries-gasp-diagnostics/audits/9

BibTeX 인용

@techreport{acceleratedindustries-acceleratedindustries-gasp-diagnostics-2026,
  author = {AcceleratedIndustries},
  title = {gasp-diagnostics security audit report (audit version 9)},
  institution = {Skillstore},
  year = {2026},
  number = {9},
  url = {https://skillstore.io/skills/acceleratedindustries-gasp-diagnostics/audits/9},
  note = {Author version unspecified}
}

CITATION.cff

cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "gasp-diagnostics security audit report (audit version 9)"
version: "unspecified"
type: report
authors:
  - name: "AcceleratedIndustries"
date-released: "2026-07-07"
url: "https://skillstore.io/skills/acceleratedindustries-gasp-diagnostics/audits/9"
identifiers:
  - type: other
    value: "skillstore:acceleratedindustries-gasp-diagnostics:audit:9"
    description: "Skillstore immutable audit report identifier"

Skillstore 점수

이 점수의 이유 증거 신뢰도: 높음

아키텍처

유지보수성

콘텐츠

커뮤니티

사양 준수

만들 수 있는 것

Evaluar una estación de trabajo lenta

Identifica si la saturación de CPU, la presión de memoria, la E/S de disco, la carga de GPU o un proceso específico explican el bajo rendimiento.

Comparar nodos de homelab o Proxmox

Obtiene métricas de varios hosts y encuentra el nodo que difiere de los patrones de recursos esperados.

Investigar la inestabilidad de un servicio

Usa unidades fallidas, historial de reinicios, resúmenes de registros y métricas de recursos para acotar las causas de fallos del servicio.

이 프롬프트를 사용해 보세요

Comprobar un host

Comprueba {hostname} con GASP y dime si está en buen estado. Céntrate primero en el problema más importante.

Analizar métricas pegadas

Analiza esta salida GASP. Resume el estado, las causas probables y los siguientes pasos seguros: {paste_metrics_here}

Comparar varios hosts

Compara {host_a}, {host_b} y {host_c}. Identifica valores atípicos, síntomas compartidos y la causa raíz más probable.

Investigar una regresión

Investiga por qué {hostname} se volvió más lento después de {change_or_time}. Correlaciona cambios recientes, líneas base, tendencias y principales consumidores de recursos.

모범 사례

Confirma el nombre de host de destino antes de permitir obtenciones por red.
Proporciona contexto de la carga de trabajo, cambios recientes y el rol esperado del host cuando estén disponibles.
Trata las recomendaciones como orientación diagnóstica y aprueba los cambios operativos por separado.

피하기

No expongas endpoints de métricas GASP a redes no confiables.
No obtengas datos de hosts desconocidos sin confirmar la autorización.
No reinicies servicios ni contenedores basándote solo en un pico de una métrica.

자주 묻는 질문

¿Qué es GASP?

GASP es un formato de monitorización que resume métricas del sistema Linux para diagnósticos asistidos por IA.

¿Puede esta skill obtener métricas en vivo?

Sí. Puede obtener métricas HTTP de un host con nombre cuando hay acceso de red y autorización del usuario.

¿Puede funcionar sin acceso de red?

Sí. Puedes pegar la salida GASP y la skill puede analizar los datos proporcionados.

¿Es seguro para sistemas de producción?

Puede ayudar en la evaluación de producción, pero los endpoints de métricas deben estar protegidos y los cambios operativos necesitan aprobación humana.

¿Ejecuta comandos de reparación automáticamente?

No. La skill proporciona recomendaciones diagnósticas y ejemplos de comandos, pero no cambia sistemas automáticamente.

¿Qué herramientas son compatibles?

El informe indica compatibilidad con Claude, Codex y Claude Code.