Laporan audit skill
karpathy Laporan audit.
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 OpenAgentSkill
Trust Score OpenAgentSkill
The Trust Score helps an agent decide whether a skill is safe enough to shortlist before installation.
Adopsi GitHub
Info62
163 star GitHub
Aktivitas star/fork
Peringatan57
163 star dan 19 fork; aktivitas issue tidak tersedia dalam metadata saat ini
Pemeliharaan terbaru
Info76
4 bulan sejak push
Kejelasan lisensi
Lulus86
MIT
Kelengkapan README/SKILL.md
Lulus86
Metadata memuat konteks penggunaan dan alur kerja yang cukup
Risiko dependensi/runtime
Peringatan54
command execution surface, credential or environment access
Ketersediaan pemasangan
Lulus92
npx skills add gaasher/Agent-Loop-Skills --skill karpathy
Keamanan perintah pemasangan
Lulus92
Jalur pemasangan paket atau runtime standar
Cakupan izin
Gagal36
secrets or environment access, shell or command execution
Bukti repositori
Lulus86
https://github.com/gaasher/Agent-Loop-Skills/tree/main/loops/karpathy
Status peninjauan
Lulus88
Data tinjauan AI tersedia
Hasil terbukti Agent
Info54
Belum ada data hasil Agent
Pemeriksaan
Tinjauan pemasangan dan adopsi
Jalur pemasangan
92
npx skills add gaasher/Agent-Loop-Skills --skill karpathy
Repositori
88
https://github.com/gaasher/Agent-Loop-Skills/tree/main/loops/karpathy
Lisensi
86
MIT
Pemeliharaan
76
4 bulan sejak push
Tinjauan AI
88
Approved with no listed issues
Kelengkapan README/SKILL.md
86
Usable description available
Risiko dependensi
54
command execution surface, credential or environment access
Keamanan perintah pemasangan
92
Jalur pemasangan paket atau runtime standar
Cakupan izin
36
secrets or environment access, shell or command execution
Aktivitas star/fork
57
163 star dan 19 fork; aktivitas issue tidak tersedia dalam metadata saat ini
Adopsi
68
163 star GitHub
Peringatan
- 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
Metode
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.
Bandingkan opsi sekitar
Skill terkait untuk diaudit berikutnya
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 Star · Laporan audit
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 Star · Laporan audit
Academic Research Skills
Academic Research Skills for Claude Code: research → write → review → revise → finalize
38K Star · Laporan audit