dstack-prototyping
Use with the dstack skill for model-serving work when the image, serving command, resources, backend/fleet choice, or service behavior is not proven. Guides task-first prototyping on real hardware, choosing fleets/backends that can reuse idle instances and caches, checking vLLM/S
Profil aset
Riset dan pekerjaan pengetahuan
Deep research, source comparison, literature review, RAG, knowledge search, and reports.
Skenario
Agent riset
I need my agent to research a topic, compare sources, and produce a concise report.
Kecocokan Agent
Claude Code + CLI + Codex
Cocok untuk Codex, Claude Code, Cursor, CLI, atau Agent khusus.
Pasang
Siap
npx skills add dstackai/dstack --skill dstack-prototyping
Pemeliharaan
Terkini
1 hari sejak push
Risiko
Perlu ditinjau
Dependency or permission surface needs review
Kualitas GitHub
2.2K
80/100 Kualitas · 72/100 Kepercayaan
Tag cakupan
Catatan ulasan
Dependency or permission surface needs review · Permission surface may require sandboxing
Kartu adopsi Agent
Kepercayaan, audit, dan kesiapan pemasangan dalam sekali lihat
Skor ini menggabungkan metadata repositori publik, sinyal ulasan OpenAgentSkill, kebaruan pemeliharaan, dan kesiapan pemasangan. Ini adalah sinyal shortlist, bukan pengganti peninjauan manusia.
Kualitas
KuatSolid option that is likely worth shortlisting for production workflows.
Kepercayaan
Hanya sandboxKandidat berguna dengan sinyal kepercayaan yang kurang atau bercampur. Gunakan di ruang kerja terisolasi hingga loop hasil membuktikan kecocokan tugas.
Audit
Perlu ditinjauTinjauan yang dapat dibaca mesin tentang kesiapan pemasangan, metadata keamanan, pemeliharaan, dan risiko adopsi.
Trust Score OpenAgentSkill v5
Tinjauan manusia sebelum pemasangan
Jalankan hanya dalam sandbox dan bandingkan alternatif terdekat sebelum digunakan untuk kerja nyata.
Star
2.2K star GitHub
Aktivitas repositori
2.2K star dan 250 fork
Pemeliharaan
1 hari sejak push
Lisensi
MPL-2.0
Pasang
npx skills add dstackai/dstack --skill dstack-prototyping
Keamanan pemasangan
Jalur pemasangan paket atau runtime standar
Cakupan izin
secrets or environment access, shell or command execution
Hasil Agent
Belum ada data hasil Agent
Dokumentasi
Konteks README/SKILL.md kuat
Ringkasan risiko
Tinjau sebelum produksi
- The skill does not explicitly advise on checking for malicious or untrusted images/modules when prototyping, though it points to official sources which mitigates risk.
- Quality score needs review
- Permission surface needs review: secrets or environment access, shell or command execution
- Dependency/runtime risk: command execution surface, credential or environment access
Kesiapan pemasangan
Jalur pemasangan tersedia
- Jalur pemasangan tersedia
- Bukti repositori tersedia
- Lisensi dinyatakan
- Belum ada bukti hasil Agent-Proven
Metadata yang dapat dibaca Agent
Data keputusan yang dapat dibaca mesin untuk skill ini.
Gunakan blok ini atau JSON tersemat untuk memutuskan apakah Agent perlu memasang skill ini, memilih alternatif, atau meminta tinjauan manusia terlebih dahulu.
Tugas yang sesuai
- alur kerja Local desktop
- Tim Claude Code
- Tim yang menghargai sinyal adopsi GitHub
- Navigate local resources
Agent yang sesuai
Keputusan pemasangan
- Perintah
- npx skills add dstackai/dstack --skill dstack-prototyping
- Kebijakan
- Blokir
- Tinjauan manusia
- Ya
Kepercayaan dan risiko
- Kepercayaan
- 64/100
- Audit
- 80/100
- Tingkat risiko
- Perlu ditinjau
Lingkar hasil
- Endpoint
- /api/agent/outcome
- ID event
- resolve
- Hasil
- 5
Perintah pemasangan
npx skills add dstackai/dstack --skill dstack-prototypingJangan gunakan ketika
- Tim yang membutuhkan SLA dengan dukungan vendor
- production agents without a repository review
- The skill does not explicitly advise on checking for malicious or untrusted images/modules when prototyping, though it points to official sources which mitigates risk.
- Petunjuk izin berisiko tinggi: Shell or command execution, Secrets or environment access
- Dependency or permission surface needs review
Skill alternatif
Last30days Skill
53.5K Star
npx skills add mvanhorn/last30days-skill -g
Skill alternatif
Academic Research Skills
38.4K Star
npx skills add Imbad0202/academic-research-skills
Skill alternatif
GPT Researcher
28.0K Star
npx skills add assafelovic/gpt-researcher
Skill alternatif
DeepResearch
19.8K Star
npx skills add Alibaba-NLP/DeepResearch
Keamanan Agent v2
40/100 · Hindari pemasangan otomatis
This skill should not be selected by an agent without explicit human security review.
Do not auto-install. Inspect the source, dependencies, and permission surface first.
Tinggi
Eksekusi shell atau perintah
Metadata skill merujuk terminal, CLI, shell, subprocess, atau alur kerja eksekusi perintah.
Sedang
Akses jaringan
Skill kemungkinan mengambil halaman jarak jauh, API, repositori, atau layanan eksternal.
Sedang
Akses sistem file
Skill dapat membaca atau menulis file proyek, dokumen, artefak yang dihasilkan, atau status workspace lokal.
Tinggi
Secrets or environment access
Skill metadata references credentials, tokens, environment variables, or secret-bearing workflows.
- Petunjuk izin berisiko tinggi: Shell or command execution, Secrets or environment access
- Dependency or permission surface needs review
Target pemasangan
Pasang skill ini di alur Agent Anda
Gunakan endpoint publik untuk mengambil perintah, checklist keamanan, prompt target, dan tautan kanonis.
OpenAgentSkill CLI
Resolve policy, run the source installer safely, and report a verified install receipt.
$ npx --yes https://github.com/Leon-Drq/openagentskill/releases/download/cli-v0.2.1/openagentskill-0.2.1.tgz install dstackai-dstack-prototypingRencana resolusi Agent
Biarkan Agent memverifikasi kecocokan sebelum memasang.
API Resolve mengembalikan skill utama, alternatif, kebijakan keamanan, catatan audit, target pemasangan, dan prompt siap pakai.
Buka JSON
/api/agent/resolve?task=Use%20dstack-prototyping%20for%20an%20agent%20workflow&agent=codex&max_risk=medium
Teks Resolve
/api/agent/resolve?task=Use%20dstack-prototyping%20for%20an%20agent%20workflow&agent=codex&max_risk=medium&format=text
Serah-terima pemasangan
/api/skills/dstackai-dstack-prototyping/install
Agent harus memeriksa
- Task fit and alternatives from Resolve API.
- Audit score, trust score, and safety policy warnings.
- Install target compatibility for Codex, Claude Code, Cursor, or CLI.
Salin prompt
Task: Use dstack-prototyping in this workspace.
Resolve first: https://www.openagentskill.com/api/agent/resolve?task=Use%20dstack-prototyping%20for%20an%20agent%20workflow&agent=codex&max_risk=medium
Review install handoff: https://www.openagentskill.com/api/skills/dstackai-dstack-prototyping/install
Install command: npx skills add dstackai/dstack --skill dstack-prototyping
Before running it, summarize audit warnings, required permissions, and the fallback skill if install is risky.Serah-terima Agent
Berikan jalur pemasangan kepada Agent, bukan direktori lain.
Gunakan endpoint publik untuk mengambil perintah, checklist keamanan, prompt target, dan tautan kanonis.
Serah-terima pemasangan
/api/skills/dstackai-dstack-prototyping/install
Format teks LLM
/api/skills/dstackai-dstack-prototyping/install?format=text
Cari alternatif
/api/skills/search?q=dstack-prototyping&limit=3
Prompt Agent
Use dstack-prototyping for this task. Review https://www.openagentskill.com/api/skills/dstackai-dstack-prototyping/install, then install with: npx skills add dstackai/dstack --skill dstack-prototypingMetadata Registry
Profil yang dapat dibaca Agent untuk pemilihan skill otomatis.
API Registry menyediakan sinyal keputusan, kepercayaan, audit, use case, dan pemasangan tanpa mengikis UI.
Manifest
/api/registry/manifest/dstackai-dstack-prototyping
Teks LLM
/api/registry/manifest/dstackai-dstack-prototyping?format=text
Alias pemasangan
/api/registry/install/dstackai-dstack-prototyping
Rekomendasikan
/api/registry/recommend?task=Use%20dstack-prototyping%20in%20an%20agent%20workflow&limit=3
Kecocokan Agent
Local desktop
Tag use case
Platform
Claude Code
Laporan audit
Perlu ditinjau · 80/100
Tinjauan yang dapat dibaca mesin tentang kesiapan pemasangan, metadata keamanan, pemeliharaan, dan risiko adopsi.
Panel keputusan Agent
Pilihan utama untuk Local desktop
Use this as a leading candidate, then validate the README and install path in your own agent stack.
Peran di stack
Pilihan utama
Kecocokan utama
Local desktop
Label kepercayaan
Siap produksi
Jalur pemasangan
Perintah siap
Gunakan saat
- alur kerja Local desktop
- Tim Claude Code
- Tim yang menghargai sinyal adopsi GitHub
Bukti
- 2,216 star GitHub
- recent repository activity
- install command or GitHub repo available
- profil kualitas 80/100
- 4 event interaksi OpenAgentSkill
tinjau dulu
- The skill does not explicitly advise on checking for malicious or untrusted images/modules when prototyping, though it points to official sources which mitigates risk.
Jalur implementasi
- 1Pasang di Agent sandbox dan jalankan satu tugas Local desktop dari awal hingga akhir.
- 2Compare output quality, latency, and failure behavior against at least one alternative.
- 3Promote it into production only after reviewing repository permissions, license, and maintenance signals.
Profil kepercayaan
Hanya sandbox
Kandidat berguna dengan sinyal kepercayaan yang kurang atau bercampur. Gunakan di ruang kerja terisolasi hingga loop hasil membuktikan kecocokan tugas.
Adopsi GitHub
Lulus2.2K star GitHub
Aktivitas star/fork
Lulus2.2K star dan 250 fork; aktivitas issue tidak tersedia dalam metadata saat ini
Pemeliharaan terbaru
Lulus1 hari sejak push
Kejelasan lisensi
LulusMPL-2.0
Sinyal positif
- Tinjauan AI disetujui
- Jalur pemasangan tersedia
- Bukti repositori tersedia
- Repositori yang baru dipelihara
- Sinyal adopsi GitHub yang bermakna
- Perintah pemasangan tidak memiliki pola berisiko tinggi yang jelas
- Loop hasil siap tetapi membutuhkan eksekusi Agent nyata pertama
Tinjau sebelum memasang
- The skill does not explicitly advise on checking for malicious or untrusted images/modules when prototyping, though it points to official sources which mitigates risk.
- Quality score needs review
- Permission surface needs review: secrets or environment access, shell or command execution
- Dependency/runtime risk: command execution surface, credential or environment access
- Permission surface: secrets or environment access, shell or command execution
- Belum ada laporan hasil Agent nyata
- Tinjauan manusia diperlukan sebelum pemasangan tanpa pengawasan
Tindakan yang disarankan
Jalankan hanya dalam sandbox dan bandingkan alternatif terdekat sebelum digunakan untuk kerja nyata.
Profil kualitas
Kuat kandidat untuk alur kerja Agent
Solid option that is likely worth shortlisting for production workflows.
Kecocokan alur kerja
Gunakan skill ini pada skenario berikut
Operate local tools
Local desktop
I need my agent to operate local files and desktop apps in a repeatable workflow.
Manage repositories
GitHub automation
I need my agent to triage GitHub issues, review pull requests, and summarize repository changes.
Operate web apps
Browser automation
I need my agent to control a browser, fill forms, and verify web app workflows.
Kecocokan alur kerja
Tambahkan ke alur kerja lengkap
Operate and verify web apps
Browser QA agent
A workflow for agents that navigate products, fill forms, take screenshots, and verify real user flows across web applications.
Find, compare, and synthesize
Research report agent
A workflow for agents that gather sources, compare claims, summarize long material, and draft useful research briefs.
Ingest, retrieve, and cite
RAG knowledge base
A workflow for document-heavy agents that ingest files, create searchable knowledge, retrieve relevant context, and answer with grounded sources.
Daftar alternatif
Bandingkan sebelum memasang
Similar skills that may fit this task.
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.
Academic Research Skills
Academic Research Skills for Claude Code: research → write → review → revise → finalize
GPT Researcher
Run autonomous deep research over web and local sources
DeepResearch
Tongyi Deep Research, the Leading Open-source Deep Research Agent
Ringkasan
--- name: dstack-prototyping description: | Use with the dstack skill for model-serving work when the image, serving command, resources, backend/fleet choice, or service behavior is not proven. Guides task-first prototyping on real hardware, choosing fleets/backends that can reuse idle instances and caches, checking vLLM/SGLang sources, and verifying the final dstack service with a model request. ---
# dstack Prototyping
Use `/dstack` for CLI commands, YAML fields, apply/attach behavior, service URLs, and other dstack syntax. This skill explains how to use dstack runs while the model-serving configuration is still unknown.
## Goal
Find a working dstack service configuration for the requested model.
Before submitting a service, use a task on real hardware to test the serving image, install/runtime assumptions, model download, cache path, command, port, launch flags, resources, env vars, backend/fleet choice, and local model request. Then submit the same configuration as a service and verify the model through the dstack service URL.
## Choose Where To Run
Pick the offer whose hardware best fits the goal at hand. Only when several offers fit comparably, choose a VM-based backend, an SSH fleet, or a Kubernetes fleet: they support idle instances and/or instance volumes, so later runs reuse the provisioned/idle instance or instance volumes for caching model weights (and possibly other writes), while container-based backends start clean on every run.
Fetch `https://dstack.ai/docs/concepts/backends.md` and classify backends from the fetched document, not from memory.
## Check Serving Sources
Check serving-framework sources early enough to choose the image, command, launch flags, resources, cache paths, request format, and expected model behavior.
For vLLM and SGLang, use these as credible sources:
- vLLM recipes and model index: `https://recipes.vllm.ai/` and `https://recipes.vllm.ai/models.json` - SGLang docs: `https://docs.sglang.io/` (fetch `/llms.txt` for the page index) - SGLang model recipes: `https://docs.sglang.io/cookbook/autoregressive/intro` - Release notes: `https://github.com/vllm-project/vllm/releases` and `https://github.com/sgl-project/sglang/releases` - Performance-loop methodology (profiling, benchmark contracts): `https://www.lmsys.org/blog/2026-07-02-agent-assisted-sglang-development`
## Use A Task Before Service
Before submitting a service, start a long-lived task:
```yaml commands: - sleep infinity ```
or an equivalent idle command.
Submit the task detached, attach or SSH into it when available, and run commands inside the live environment. Test the image, installs, model download and cache path, serving command, port, launch flags, local model request, and expected model behavior.
When starting a long-running command in the background from a non-interactive SSH command, use `nohup`, redirect stdin from `/dev/null`, and redirect stdout/stderr to a log file so the SSH command returns while the process keeps running. For example (the command can be any long-running command):
```shell nohup vllm serve ... </dev/null > /tmp/vllm.log 2>&1 & ```
If the image, hardware choice, or major install path changes, submit another task so the changed setup is tested before service verification.
Do not move to a service after checking only GPU visibility, imports, logs, or a health endpoint. Start the server inside the task and send a request that uses the requested model. For a chat or reasoning model, check the response behavior the endpoint is expected to support, such as reasoning output when that model is supposed to expose it.
Follow `/dstack` structured status guidance when polling task or service status. After requesting a task or service stop before another submission, wait until that run reaches a terminal status. This allows dstack to reuse its instance or instance volumes when available.
## Verify As A Service
Submit the service after the task has verified the configuration: image, command, port, resources, env vars, cache mounts if used, backend/fleet choice, and model request.
Use the service as a duplicate check of the same configuration under dstack service runtime. The model request that worked locally in the task must also work through the dstack service URL.
If service verification fails because the image, install, model download, command, resources, cache, or model behavior needs to change, go back to a task. If the tested serving setup is still right and only the dstack service configuration is wrong, fix the configuration and submit the service again.
Detail teknis
- Versi
- 1.0.0
- Lisensi
- MPL-2.0
- Pembaruan terakhir
- 21 Agu 2026
- Diterbitkan
- 21 Agu 2026
Ringkasan keputusan
Pilihan utama
2,216 star GitHub
Audit
Tinjauan pemasangan
Tinjauan pemasangan dan adopsi
- Keamanan
- 70/100
- Pemeliharaan
- 100/100
- Pasang
- 92/100
Bukti tervalidasi Agent
Bukti tervalidasi Agent
Laporan hasil setelah resolve, tinjau, pasang, dan satu eksekusi terbatas.
- Tingkat sukses
- —
- Kegagalan terbaru
- —
- Hasil
- 0
- Kualitas output
- —
- Gagal
- 0
- Tidak relevan
- 0
- Pemasangan
- 0
- Diblokir risiko
- 0
- Perlu penyiapan
- 0
- Produksi
- 0
Belum ada data hasil Agent. Eksekusi pertama dapat melaporkan keberhasilan, kebutuhan setup, blok risiko, kegagalan, atau tidak relevan melalui /api/agent/outcome.
Pasang
Tambahkan ke alur Agent
Gratis dan sumber terbuka. Tinjau laporan sebelum memasang pada Agent produksi.
Siklus pertumbuhan
Kit berbagi
Draf berbasis skenario untuk dstack-prototyping, siap untuk posting manual di X.
dstack-prototyping: Use with the dstack skill for model-serving work when the image, serving command, resources,... 2.2K stars https://www.openagentskill.com/skills/dstackai-dstack-prototyping?ref=x
Balasan opsional dengan perintah pemasangan
Listing + install path for dstack-prototyping: https://www.openagentskill.com/skills/dstackai-dstack-prototyping?ref=x Install: npx skills add dstackai/dstack --skill dstack-prototyping
Sumber listing
Diindeks Registry
Listing ini diindeks dari sumber publik dan belum ditandai resmi hingga klaim pemelihara disetujui.
- Kreator
- dstackai
- Sumber
- dstackai/dstack
- Diindeks oleh
- Indeks komunitas OpenAgentSkill
Atribusi menautkan ke repositori publik atau profil kreator. Kreator dapat mengklaim listing untuk memperbarui sinyal kepemilikan.
Klaim skill iniKlaim pemilik
Klaim listing skill ini
Listing Diindeks Registry ini dikaitkan dengan dstackai, tetapi belum ditandai resmi. Klaim untuk menambahkan sinyal pemilik terverifikasi dan membuat pembaruan peluncuran, pemasangan, serta audit berikutnya lebih tepercaya.
Kit backlink kreator
Tambahkan badge bukti ke README Anda
Tampilkan listing kanonis, sinyal kepercayaan dan audit saat ini, serta bukti Agent-Proven nyata di tempat pengembang mengevaluasi repositori.
[](https://www.openagentskill.com/skills/dstackai-dstack-prototyping)
[](https://www.openagentskill.com/skills/dstackai-dstack-prototyping)
[](https://www.openagentskill.com/skills/dstackai-dstack-prototyping/audit)
[](https://www.openagentskill.com/skills/dstackai-dstack-prototyping)Penulis
dstackai
@dstackai
Tag
Kecocokan platform
Sinyal kesehatan
- Star GitHub
- 2.2K
- Skor kualitas
- 47/100
- Push GitHub terakhir
- 21 Agu 2026
- Petunjuk framework
- Tidak diketahui
- Tampilan OpenAgentSkill
- 4
- Salinan pemasangan
- 0
- Klik keluar
- 0
Sinyal komunitas
Bagikan apakah skill ini bermanfaat untuk alur kerja Agent Anda. Masukan gabungan meningkatkan peringkat dari waktu ke waktu.
Kepercayaan & keamanan
Hanya sandbox
- Adopsi GitHub2.2K star GitHubLulus
- Aktivitas star/fork2.2K star dan 250 fork; aktivitas issue tidak tersedia dalam metadata saat iniLulus
- Pemeliharaan terbaru1 hari sejak pushLulus
- Kejelasan lisensiMPL-2.0Lulus
- Kelengkapan README/SKILL.mdMetadata memuat konteks penggunaan dan alur kerja yang cukupLulus
- Risiko dependensi/runtimecommand execution surface, credential or environment accessPerbaiki
Skill terkait
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.
53.5K StarAcademic Research Skills
Academic Research Skills for Claude Code: research → write → review → revise → finalize
38.4K StarGPT Researcher
Run autonomous deep research over web and local sources
28.0K StarDeepResearch
Tongyi Deep Research, the Leading Open-source Deep Research Agent
19.8K Star