Menu

#493 by GitHub stars

dstackai/dstack

A unified orchestration layer for heterogeneous AI compute. It standardizes how to manage compute and run training and inference on GPU clouds, Kubernetes, VMs, or bare-metal clusters.

CodexDevOpsEducation

History

Review weekly changes to SKILL.md before updating. Dates mark the weeks in which we observed each version; checks apply to the recorded commits.

View weekly changes

skills/dstack/SKILL.md

Week of Nov 2, 2026 · +4 / −4 lines

Before Network or broad shell → After Network or broad shell

be2b22c8056f → 4c3b43666e93

- 568 - [Overview](https://dstack.ai/docs/overview.md)
+ 568 - [Overview](https://dstack.ai/docs/index.md)
- 583 - [AMD](https://dstack.ai/examples/accelerators/amd/index.md)
- 584 - [Google TPU](https://dstack.ai/examples/accelerators/tpu/index.md)
- 585 - [Tenstorrent](https://dstack.ai/examples/accelerators/tenstorrent/index.md)
+ 583 - [AMD](https://dstack.ai/docs/examples/accelerators/amd.md)
+ 584 - [Google TPU](https://dstack.ai/docs/examples/accelerators/tpu.md)
+ 585 - [Tenstorrent](https://dstack.ai/docs/examples/accelerators/tenstorrent.md)

Compare on GitHub

Week of Oct 26, 2026 · Line counts unavailable

Before Not checked yet → After Network or broad shell

First observation → be2b22c8056f

No earlier version was observed for comparison.

Compare on GitHub

skills/dstack-presets/SKILL.md

Week of Nov 2, 2026 · +0 / −0 lines

Before Instructions only → After Network or broad shell

be2b22c8056f → 4c3b43666e93

Compare on GitHub

Week of Oct 26, 2026 · Line counts unavailable

Before Not checked yet → After Instructions only

First observation → be2b22c8056f

No earlier version was observed for comparison.

Compare on GitHub

skills/dstack-prototyping/SKILL.md

Week of Nov 2, 2026 · +0 / −0 lines

Before Network or broad shell → After Network or broad shell

be2b22c8056f → 4c3b43666e93

Compare on GitHub

Week of Oct 26, 2026 · Line counts unavailable

Before Not checked yet → After Network or broad shell

First observation → be2b22c8056f

No earlier version was observed for comparison.

Compare on GitHub

View on GitHubAdd to my skills

Verified commit 4c3b43666e936e066708d16f6204f133c01f36b1

git clone --filter=blob:none --no-checkout https://github.com/dstackai/dstack.git .hotskills/dstackai-dstack && git -C .hotskills/dstackai-dstack checkout --detach 4c3b43666e936e066708d16f6204f133c01f36b1 # Copy each selected skill directory from .hotskills/dstackai-dstack into .agents/skills/<skill-name>/
2.3kGitHub stars
8stars in 7 days
3skills in this repository
75commits in 30 days

Heat

5.7
  • Star growth, last 7 daysAhead of 66% of tracked repositories
  • Star growth, last 24 hoursAhead of 68% of tracked repositories
  • Rank movement, last 7 daysAhead of 43% of tracked repositories
  • Commits, last 30 daysAhead of 93% of tracked repositories
  • Accounts asking about it through MCP, last 7 daysNot available
  • Hot votes this weekAhead of 0% of tracked repositories

Heat never changes the order by stars, and sponsorship never touches it. How heat works

Stars over time

Jul 31, 2026Nov 6, 20262,304

Individual skills in this repository

How it works

dstack

When your agent uses it

dstack is an open-source control plane for GPU provisioning and orchestration across GPU clouds, Kubernetes, and on-prem clusters.

What it loads into context

Frontmatter

34 tokens

Always available to the agent

SKILL.md

6,152 tokens

When the skill is activated

Reference documents

0 tokens

Loaded on request

Token counts are estimates. Scripts run separately; they are not counted as reference context.

The playbook

  1. Overview
  2. How it works
  3. Quick agent flow (detached runs)
  4. Agent execution guidelines
  5. Configuration types
  6. Essential CLI commands
  7. Troubleshooting
  8. Additional resources

Files

  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
  • SKILL.mdcore24,989 bytes
Read SKILL.md

# dstack

## Overview

`dstack` provisions and orchestrates workloads across GPU clouds, Kubernetes, and on-prem via fleets.

**When to use this skill:**
- Running or managing dev environments, tasks, or services on dstack
- Creating, editing, or applying `*.dstack.yml` configurations
- Managing fleets, volumes, gateways, and checking available offers

## How it works

`dstack` operates through three core components:
View the complete source on GitHub

Skills included

  • dstack
  • dstack-presets
  • dstack-prototyping

Repository facts

License
MPL-2.0
Created
Jan 4, 2022
Last pushed
Nov 5, 2026
Forks
272
Open issues
60
Project activity
Active

Rank over the last 12 weeks

Now #493 · best #361 · down 70 in 30 days

  1. #485 from Oct 30 to Oct 31
  2. #486 from Nov 1 to Nov 2
  3. #488 on Nov 3
  4. #492 on Nov 4
  5. #491 on Nov 5
  6. #493 since Nov 6
56 earlier periods
  1. #355 on Aug 1
  2. #345 on Aug 7
  3. #347 on Aug 8
  4. #359 on Aug 10
  5. #352 on Aug 12
  6. #357 on Aug 13
  7. #356 on Aug 14
  8. #361 on Aug 17
  9. #369 on Aug 18
  10. #366 on Aug 19
  11. #368 on Aug 21
  12. #364 on Aug 22
  13. #375 on Aug 25
  14. #374 on Aug 26
  15. #377 on Aug 27
  16. #378 on Aug 28
  17. #372 on Aug 29
  18. #376 on Aug 31
  19. #386 on Sep 1
  20. #380 on Sep 2
  21. #385 on Sep 3
  22. #376 on Sep 4
  23. #380 on Sep 5
  24. #386 on Sep 7
  25. #389 from Sep 8 to Sep 9
  26. #388 on Sep 11
  27. #392 on Sep 12
  28. #397 on Sep 14
  29. #401 on Sep 15
  30. #402 on Sep 16
  31. #396 on Sep 17
  32. #401 on Sep 20
  33. #397 on Sep 21
  34. #408 on Sep 22
  35. #399 on Sep 23
  36. #396 on Sep 26
  37. #405 on Sep 27
  38. #412 on Sep 28
  39. #418 on Sep 30
  40. #416 on Oct 1
  41. #417 on Oct 2
  42. #413 on Oct 5
  43. #417 on Oct 6
  44. #423 on Oct 7
  45. #436 on Oct 9
  46. #423 on Oct 10
  47. #436 on Oct 12
  48. #434 on Oct 13
  49. #435 on Oct 14
  50. #450 on Oct 19
  51. #455 on Oct 21
  52. #453 on Oct 23
  53. #442 on Oct 24
  54. #473 on Oct 25
  55. #481 on Oct 28
  56. #483 on Oct 29

Before you install

Network or broad shell

Automated signals, not a guarantee

Checked Nov 4, 2026

Not fully analyzed: some files were too long, unreadable or missing. This never lowers the level.

Show 41 notes
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
  • skills/dstack/SKILL.md:1 Analysis incomplete: Referenced executable target unavailable; not analyzed
Show all 49 signals by file

skills/dstack-presets/SKILL.md

  • Network accessline 27
    Follow the [presets documentation](https://dstack.ai/docs/concepts/presets.md).
  • Network accessline 29
    [Configuration reference](https://dstack.ai/docs/reference/dstack.yml/preset.md) | [CLI reference](https://dstack.ai/docs/reference/cli/dstack/preset.md)

skills/dstack-prototyping/SKILL.md

  • Network accessline 27
    Fetch `https://dstack.ai/docs/concepts/backends.md` and classify backends
  • Network accessline 40
    - 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
  • Network accessline 41
    `https://recipes.vllm.ai/models.json` - SGLang docs: `https://docs.sglang.io/` (fetch `/llms.txt` for the page   index)
  • Network accessline 42
    - SGLang docs: `https://docs.sglang.io/` (fetch `/llms.txt` for the page
  • Network accessline 43
    index) - SGLang model recipes: `https://docs.sglang.io/cookbook/autoregressive/intro` - Release notes: `https://github.com/vllm-project/vllm/releases` and
  • Network accessline 44
    - SGLang model recipes: `https://docs.sglang.io/cookbook/autoregressive/intro`
  • Network accessline 45
    - Release notes: `https://github.com/vllm-project/vllm/releases` and
  • Network accessline 46
    `https://github.com/sgl-project/sglang/releases`
  • Network accessline 48
    `https://www.lmsys.org/blog/2026-07-02-agent-assisted-sglang-development`
  • Network accessline 120
    - When testing inference, call the router endpoint, not the workers directly   (unless you want to test if they are alive). - Look for "Prototyping services" in `https://dstack.ai/docs/concepts/tasks.md`
  • Network accessline 121
    (unless you want to test if they are alive). - Look for "Prototyping services" in `https://dstack.ai/docs/concepts/tasks.md`   and "Router" in `https://dstack.ai/docs/concepts/services.md`.
  • Network accessline 122
    - Look for "Prototyping services" in `https://dstack.ai/docs/concepts/tasks.md`
  • Network accessline 123
    and "Router" in `https://dstack.ai/docs/concepts/services.md`.
  • Network accessline 130
    separate groups, and the fleet needs an interconnect (`placement: cluster`). - Look for "Node groups" and "PD disaggregation" in   `https://dstack.ai/docs/concepts/tasks.md` and "Replica groups" and
  • Network accessline 131
    - Look for "Node groups" and "PD disaggregation" in   `https://dstack.ai/docs/concepts/tasks.md` and "Replica groups" and   "PD disaggregation" in `https://dstack.ai/docs/concepts/services.md`.
  • Network accessline 132
    `https://dstack.ai/docs/concepts/tasks.md` and "Replica groups" and
  • Network accessline 133
    "PD disaggregation" in `https://dstack.ai/docs/concepts/services.md`.

skills/dstack/SKILL.md

  • Access to secretsline 117
    [Credential-related source omitted]
  • Access to secretsline 143
    [Credential-related source omitted]
  • Access to secretsline 173
    [Credential-related source omitted]
  • Network accessline 208
    [Concept documentation](https://dstack.ai/docs/concepts/dev-environments.md) | [Configuration reference](https://dstack.ai/docs/reference/dstack.yml/dev-environment.md)
  • Network accessline 236
    [Concept documentation](https://dstack.ai/docs/concepts/tasks.md) | [Configuration reference](https://dstack.ai/docs/reference/dstack.yml/task.md)
  • Network accessline 268
    curl -sS -X POST "https://<run name>.<gateway domain>/v1/chat/completions" \
  • Network accessline 276
    [Concept documentation](https://dstack.ai/docs/concepts/services.md) | [Configuration reference](https://dstack.ai/docs/reference/dstack.yml/service.md)
  • Access to secretsline 305
    [Credential-related source omitted]
  • Network accessline 311
    [Concept documentation](https://dstack.ai/docs/concepts/fleets.md) | [Configuration reference](https://dstack.ai/docs/reference/dstack.yml/fleet.md)
  • Network accessline 342
    [Concept documentation](https://dstack.ai/docs/concepts/volumes.md) | [Configuration reference](https://dstack.ai/docs/reference/dstack.yml/volume.md)
  • Network accessline 359
    [Concept documentation](https://dstack.ai/docs/concepts/gateways.md) | [Configuration reference](https://dstack.ai/docs/reference/dstack.yml/gateway.md)
  • Network accessline 561
    1. Display the full error message unchanged 2. Do NOT retry the same command without addressing the error 3. Refer to the [Troubleshooting guide](https://dstack.ai/docs/guides/troubleshooting.md) for guidance
  • Network accessline 563
    3. Refer to the [Troubleshooting guide](https://dstack.ai/docs/guides/troubleshooting.md) for guidance
  • Network accessline 567
    **Core documentation:** - [Overview](https://dstack.ai/docs/index.md)
  • Network accessline 568
    - [Overview](https://dstack.ai/docs/index.md)
  • Network accessline 569
    - [Installation](https://dstack.ai/docs/installation.md)
  • Network accessline 570
    - [Quickstart](https://dstack.ai/docs/quickstart.md)
  • Network accessline 572
    **Additional concepts:** - [Secrets](https://dstack.ai/docs/concepts/secrets.md)
  • Network accessline 573
    - [Secrets](https://dstack.ai/docs/concepts/secrets.md)
  • Network accessline 574
    - [Projects](https://dstack.ai/docs/concepts/projects.md)
  • Network accessline 575
    - [Metrics](https://dstack.ai/docs/concepts/metrics.md)
  • Network accessline 576
    - [Events](https://dstack.ai/docs/concepts/events.md)
  • Network accessline 578
    **Guides:** - [CLI & API](https://dstack.ai/docs/guides/cli-api.md)
  • Network accessline 579
    - [CLI & API](https://dstack.ai/docs/guides/cli-api.md)
  • Network accessline 580
    - [Server deployment](https://dstack.ai/docs/guides/server-deployment.md)
  • Network accessline 582
    **Accelerator-specific examples:** - [AMD](https://dstack.ai/docs/examples/accelerators/amd.md)
  • Network accessline 583
    - [AMD](https://dstack.ai/docs/examples/accelerators/amd.md)
  • Network accessline 584
    - [Google TPU](https://dstack.ai/docs/examples/accelerators/tpu.md)
  • Network accessline 585
    - [Tenstorrent](https://dstack.ai/docs/examples/accelerators/tenstorrent.md)
  • Network accessline 587
    **Full documentation:** https://dstack.ai/llms-full.txt

How the signals fit together

Facts from a static read of the files at the pinned commit. They describe what the code does, not why.

Some files could not be read; the facts below still hold.

Combined signals: 1 fact
  • Reads SSH keys, and the same file sends data over the network or runs a command (line 268)dstack/SKILL.md:305High confidence
Signal confidence: 49 facts

Works with

Agent compatibility, from the files in this repository
AgentStatusEvidenceInstall to
Claude CodeCompatibleStandard SKILL.md skills/dstack/SKILL.md.claude/skills/<skill-name>/
CodexNativeManifest AGENTS.md.agents/skills/<skill-name>/
CursorCompatibleStandard SKILL.md skills/dstack/SKILL.md.cursor/skills/<skill-name>/
Gemini CLICompatibleStandard SKILL.md skills/dstack/SKILL.md.gemini/skills/<skill-name>/
GitHub CopilotCompatibleStandard SKILL.md skills/dstack/SKILL.md.github/skills/<skill-name>/
OpenCodeCompatibleStandard SKILL.md skills/dstack/SKILL.md.opencode/skills/<skill-name>/
  • Native The repository ships a manifest, folder or frontmatter field for this agent.
  • Compatible A standard SKILL.md that this agent can read.
  • Unknown No evidence either way.

Derived automatically from paths, manifests and frontmatter. Not tested on each agent.

You might also use

metalbear-co/mirrord

Run any process, on your machine or in an AI agent's environment, as if it were a pod in your Kubernetes cluster: real env vars, DNS, network, traffic.

5.4kGitHub stars

rohitg00/kubectl-mcp-server

Published in CNCF Landscape: A MCP server for Kubernetes.

968GitHub stars

aai-labs/agent-barn

Open-source control plane for AI agents on your own Kubernetes infrastructure, with per-agent costs, access controls, and audit logs.

111GitHub stars

edtechre/pybroker

Algorithmic Trading in Python with Machine Learning

3.7kGitHub stars

SharpAI/DeepCamera

Open-Source AI Camera Skills Platform, AI NVR & CCTV Surveillance. Local VLM video analysis with Qwen, DeepSeek, SmolVLM, LLaVA, YOLO26. LLM-powered agentic security camera agent …

3.1kGitHub stars

overmind-core/overmind

The platform for continuously improving AI agents.

751GitHub stars

Explore alternatives and comparisons

Add this badge to your README

It updates every night and links to this page.

dstackai/dstack: #493 on hotskills, 2.3k stars

[![dstackai/dstack on hotskills](https://hotskills.dev/badge/dstackai-dstack.svg)](https://hotskills.dev/skill/dstackai-dstack/)