FreePublished on GitHub

deep-analyze

Router for DeepAnalyze agentic data-science workflows.

The English summary is being prepared. Showing the original description.

VectorSpaceLab/AREX-Skill

Works withDocker OpenAI

See how to install

Files included (36)

SKILL.md3.8 KB
references/install-and-environment.md2.3 KB
references/repo-provenance.md1.1 KB
references/repo-routing-metadata.json561 B
references/troubleshooting.md3.5 KB
scripts/check_deepanalyze_environment.py8.9 KB
sub-skills/api-and-clients/references/api-reference.md4.5 KB
sub-skills/api-and-clients/references/case-study-patterns.md2.8 KB
sub-skills/api-and-clients/references/programmatic-usage.md2.9 KB
sub-skills/api-and-clients/references/troubleshooting.md4.8 KB
sub-skills/api-and-clients/scripts/api_client_smoke.py12.3 KB
sub-skills/api-and-clients/scripts/deepanalyze_vllm_smoke.py4.8 KB
sub-skills/api-and-clients/scripts/mock_vllm_server.py8.2 KB
sub-skills/api-and-clients/scripts/openai_client_smoke.py10.6 KB
sub-skills/api-and-clients/SKILL.md1.6 KB
sub-skills/interactive-frontends/references/cli-reference.md2.9 KB
sub-skills/interactive-frontends/references/jupyter-frontend.md2.6 KB
sub-skills/interactive-frontends/references/troubleshooting.md4.2 KB
sub-skills/interactive-frontends/references/webui-v2.md5.7 KB
sub-skills/interactive-frontends/scripts/check_webui_prereqs.py14.0 KB
sub-skills/interactive-frontends/SKILL.md2.2 KB
sub-skills/model-serving/references/cloud-api-usage.md1.6 KB
sub-skills/model-serving/references/model-customization.md3.1 KB
sub-skills/model-serving/references/troubleshooting.md3.0 KB
sub-skills/model-serving/references/vllm-and-docker.md4.4 KB
sub-skills/model-serving/scripts/quantization_command_builder.py3.1 KB
sub-skills/model-serving/scripts/token_tag_plan.py3.0 KB
sub-skills/model-serving/scripts/vllm_command_builder.py6.7 KB
sub-skills/model-serving/SKILL.md2.0 KB
sub-skills/training-and-evaluation/references/benchmark-playgrounds.md9.3 KB
sub-skills/training-and-evaluation/references/contribution-and-case-studies.md3.3 KB
sub-skills/training-and-evaluation/references/training-recipes.md10.0 KB
sub-skills/training-and-evaluation/references/troubleshooting.md7.2 KB
sub-skills/training-and-evaluation/scripts/benchmark_command_builder.py18.2 KB
sub-skills/training-and-evaluation/scripts/render_training_command.py15.4 KB
sub-skills/training-and-evaluation/SKILL.md2.8 KB

SKILL.md (original)

Review the instructions the agent will receive before you install.

DeepAnalyze

DeepAnalyze is an agentic data-science repo built around a local reasoning loop, an OpenAI-compatible API server, browser and notebook frontends, model-serving helpers, and official training/evaluation recipes.

Use this router when the request names DeepAnalyze, DeepAnalyze-8B, the OpenAI-compatible API, the browser demo, the CLI, Jupyter, vLLM serving, quantization, tokenizer customization, SFT/RL recipes, or the bundled benchmark playgrounds.

If the checkout may be stale, consult references/repo-provenance.md before trusting any route or claim.

Start here

If you are setting up or checking an environment, read references/install-and-environment.md.
If something fails, read references/troubleshooting.md.
For a read-only health check, run scripts/check_deepanalyze_environment.py against a DeepAnalyze checkout.
Then route to the sub-skill that owns the task surface.

Route map

Task family	Read first	Why
Programmatic agent loop, workspace execution, file-aware client flows	`sub-skills/api-and-clients/SKILL.md`	Covers `DeepAnalyzeVLLM`, OpenAI-style clients, file/thread semantics, and API smoke scripts.
CLI, browser WebUI v2, or Jupyter frontend	`sub-skills/interactive-frontends/SKILL.md`	Covers the terminal client, workspace UI, Docker execution mode, and Jupyter MCP flow.
Model download, vLLM launch, quantization, or tokenizer tag extension	`sub-skills/model-serving/SKILL.md`	Covers the memory table, dry-run vLLM commands, Docker deployment, and model customization.
SFT, RL, benchmark runs, or case-study contribution	`sub-skills/training-and-evaluation/SKILL.md`	Covers the official DeepAnalyze training recipes and benchmark playgrounds.

What this router should answer

How to call DeepAnalyze from Python or through an OpenAI-compatible client.
How to upload files, keep a thread workspace, and interpret generated files.
How to launch or preflight the browser and notebook experiences.
How to choose a DeepAnalyze vLLM command from GPU memory and context requirements.
How to plan tag-extension, quantization, SFT, RL, or benchmark commands without running placeholder-heavy jobs blindly.

Minimal read-only check

If you need a quick confidence check without starting services, use the bundled checker:

python scripts/check_deepanalyze_environment.py --repo-root <DeepAnalyze checkout>

That checker verifies the core Python imports, selected source-file compilation, DeepAnalyzeVLLM, API TestClient behavior, and WebUI v2 TestClient behavior.

Route away from this root when

You already know the exact sub-skill and only need its detailed reference.
The task is only about the older legacy browser demo, unless the user explicitly asks for it.
The task is about a different package's agent framework, model server, or benchmark harness.

Preferred mental model

api-and-clients = "How do I use DeepAnalyze programmatically or test the API?"
interactive-frontends = "How do I use the CLI, browser demo, or notebook UI?"
model-serving = "How do I size, serve, or customize DeepAnalyze-8B?"
training-and-evaluation = "How do I dry-run training, RL, or benchmark jobs?"

When a request spans multiple areas, pick the owner of the primary workflow first and then follow its cross-links.

Reviews

No reviews yet. Tell others what you thought.

Comments

No login needed. Share how it worked for you, ask a question, or note where it did not run.

No comments yet. Be the first to write one.

More skills from this repository

3d-resnets-pytorch

Free

Routes 3D ResNets PyTorch video action-recognition workflows across training, inference, and data preparation.

The English summary is being prepared. Showing the original description.

VectorSpaceLab/AREX-Skill338Updated Sep 3, 2026

3ddfa

Free

Guide 3DDFA Python inference, geometry rendering, training/evaluation, and optional C++ ONNX workflows for 3D dense face alignment.

The English summary is being prepared. Showing the original description.

VectorSpaceLab/AREX-Skill338Updated Sep 3, 2026

3ddfa-v2

Free

Routes 3DDFA_V2 face-alignment setup, still-image demos, video tracking, and ONNX benchmarking workflows.

The English summary is being prepared. Showing the original description.

VectorSpaceLab/AREX-Skill338Updated Sep 3, 2026

ab3dmot

Free

Operate AB3DMOT 3D multi-object tracking workflows for KITTI and nuScenes data, tracking, evaluation, and visualization.

The English summary is being prepared. Showing the original description.

VectorSpaceLab/AREX-Skill338Updated Sep 3, 2026

accelerate

Free

Use Hugging Face Accelerate for PyTorch training-loop migration, distributed launch/configuration, DeepSpeed/FSDP/TPU backend setup, big-model inference/offload, checkpointing, tracking, and troubleshooting.

The English summary is being prepared. Showing the original description.

VectorSpaceLab/AREX-Skill338Updated Sep 3, 2026

acme

Free

Route Acme reinforcement-learning framework tasks across core loops, replay/data, JAX agents, and TensorFlow/Sonnet agents.

The English summary is being prepared. Showing the original description.

VectorSpaceLab/AREX-Skill338Updated Sep 3, 2026

See all skills by VectorSpaceLab

Report a problem with this skill

Install

Install with a command

npx skills add https://github.com/VectorSpaceLab/AREX-Skill --skill deep-analyze

Paste it into your terminal and run it (requires Node.js). You can choose which agents to install it for, such as Claude Code or Codex.

Ask your agent

Please install the following Agent Skill.
https://github.com/VectorSpaceLab/AREX-Skill/tree/main/skills/repositories/repo-skills/deep-analyze

First review SKILL.md and the bundled files and confirm there is nothing suspicious (such as sending data externally or reading credentials). Then place it in ~/.claude/skills/deep-analyze/ for Claude Code, or ~/.agents/skills/deep-analyze/ for Codex.

Send this text to Claude Code or Codex as is. It checks the contents before installing.

View source on GitHub

Repository: VectorSpaceLab/AREX-Skill
Stars: 338
Forks: 26
Category: Data & AI
License: Apache-2.0
Original language: English
Last updated: Sep 3, 2026

A skill is a set of instructions and scripts for an agent. cccskills does not guarantee that skills are safe. Review the contents before you install.