Informe de auditoría del skill
karpathy Informe de auditoría.
Use when the user wants the LLM to do its own ML research: a fully-autonomous loop that hacks the training code, runs it, and keeps changes that lower a single scalar metric (e.g. val_bpb). One agent proposes one change at a time, runs training in the user's env, keeps it only if the metric improves (advancing a git branch) else reverts, and loops forever until the human interrupts. A faithful adaptation of Karpathy's autoresearch. Not for the analysis-first variant that profiles before editing (that is ml-autoresearch), and not for a budgeted, plateau-stopping refactor.
Trust Score de OpenAgentSkill
Trust Score de OpenAgentSkill
The Trust Score helps an agent decide whether a skill is safe enough to shortlist before installation.
Adopción en GitHub
Info62
163 estrellas de GitHub
Actividad de stars/forks
Advertencia57
163 estrellas y 19 forks; la actividad de issues no está disponible en los metadatos actuales
Mantenimiento reciente
Info76
4 meses desde el último push
Claridad de licencia
Aprobado86
MIT
Completitud de README/SKILL.md
Aprobado86
Los metadatos incluyen suficiente contexto de uso y flujo de trabajo
Riesgo de dependencias/runtime
Advertencia54
command execution surface, credential or environment access
Disponibilidad de instalación
Aprobado92
npx skills add gaasher/Agent-Loop-Skills --skill karpathy
Seguridad del comando de instalación
Aprobado92
Ruta estándar de paquete o instalación en tiempo de ejecución
Superficie de permisos
Fallido36
secrets or environment access, shell or command execution
Evidencia del repositorio
Aprobado86
https://github.com/gaasher/Agent-Loop-Skills/tree/main/loops/karpathy
Estado de revisión
Aprobado88
Hay datos de revisión por IA disponibles
Resultados comprobados por Agent
Info54
Aún no hay datos de resultados del Agent
Comprobaciones
Revisión de instalación y adopción
Ruta de instalación
92
npx skills add gaasher/Agent-Loop-Skills --skill karpathy
Repositorio
88
https://github.com/gaasher/Agent-Loop-Skills/tree/main/loops/karpathy
Licencia
86
MIT
Mantenimiento
76
4 meses desde el último push
Revisión por IA
88
Approved with no listed issues
Completitud de README/SKILL.md
86
Usable description available
Riesgo de dependencias
54
command execution surface, credential or environment access
Seguridad del comando de instalación
92
Ruta estándar de paquete o instalación en tiempo de ejecución
Superficie de permisos
36
secrets or environment access, shell or command execution
Actividad de stars/forks
57
163 estrellas y 19 forks; la actividad de issues no está disponible en los metadatos actuales
Adopción
68
163 estrellas de GitHub
Advertencias
- Dependency or permission surface needs review
- Permission surface may require sandboxing
- Quality score needs review
- Permission surface needs review: secrets or environment access, shell or command execution
- Stars/forks activity: 163 stars, 19 forks; issue activity unavailable in current metadata
- Dependency/runtime risk: command execution surface, credential or environment access
- Permission surface: secrets or environment access, shell or command execution
Método
This report combines public metadata, AI review output, repository freshness, install readiness, OpenAgentSkill events, quality scoring, trust checks, and the agent safety gate. It is not a full source-code security review.
Comparar opciones cercanas
Skills relacionados para auditar después
mono-color
Generate original one-ink or controlled two-ink editorial images from any theme, sentence, article idea, object, or reference photo. Always use this skill when the user asks for 单色海报、双色印刷、单色调视觉、蓝色/绿色孔版印刷、risograph、网点照片、复古或当代编辑排版、zine poster, monochrome editorial poster, duotone print, or asks to use the mono-color style. It uses an adaptive white, gray, or pale-beige substrate, no more than two printing inks, active negative space, terse human language, and strong serif/grotesk/mono typography without making retro styling the default or copying a source composition, wording, logo, or artwork. Produce both the final generation prompt and the generated raster image unless the user explicitly asks for prompt only.
1.9K Estrellas · Informe de auditoría
Last30days Skill
Research the last 30 days across Reddit, X, YouTube, Hacker News, Polymarket, GitHub, and the web, then synthesize a grounded brief for an AI agent.
64K Estrellas · Informe de auditoría
Academic Research Skills
Academic Research Skills for Claude Code: research → write → review → revise → finalize
38K Estrellas · Informe de auditoría