Skip to content
Back to skills

Tools And Demos

ASecurity

"Routes detrex demo, analysis, visualization, benchmark planning,

  • 247 stars
  • 0 votes
  • 0 copies
  • 2 views
  • Added September 8, 2026
toolsshell

Works with

  • cli

Security analysis

A100/100

Pro scans all 5 files and shows the line behind each finding

Scanned September 8, 2026

npx -y skills add VectorSpaceLab/AREX-Skill --skill tools-and-demos --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Tools And Demos?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Tools And Demos
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/vectorspacelab-tools-and-demos/badge)](https://www.skillsdirectory.com/skills/vectorspacelab-tools-and-demos)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: tools-and-demos
description: "Routes detrex demo, analysis, visualization, benchmark planning,
  and MOT demo cautions safely."
disable-model-invocation: true
metadata:
  disco-role: operating
license: Apache 2.0
---

# tools-and-demos

Use this sub-skill when you need to build or choose a safe command for:

- image, video, or webcam demo runs
- model analysis such as FLOPs, activations, parameters, or structure
- dataset or prediction visualization
- benchmark command construction without execution
- project-specific MOT demo cautions and routing

Do not use this sub-skill for:

- training or evaluation launchers
- model zoo selection or checkpoint conversion
- package installation or CUDA extension repair

## Read first

- `references/tools-reference.md` for the CLI surface and input/output rules.
- `references/demo-workflows.md` for workflow selection and command-building choices.
- `references/troubleshooting.md` for checkpoint, codec, dataset, and dependency failures.
- `scripts/build_tool_command.py` for a safe command plan with no execution.

## Skill-owned script

- `scripts/build_tool_command.py` — prints a shell command or JSON plan for demo, analysis, visualization, or benchmark workflows. It does not download weights, scan datasets, or run a benchmark by itself.

## Typical workflow

1. Pick the workflow and confirm whether you want a command plan or a runnable command.
2. Use the bundled command builder to assemble explicit config, checkpoint, input, and output arguments.
3. Check the workflow notes for checkpoint responsibility, output handling, and MOT cautions.
4. Run the printed command only after the required files and dataset registration are ready.

## Routing notes

- Use the generic demo route for independent images or videos.
- Use the visualization routes for dataset inspection or saved prediction JSON.
- Use the analysis route when you need parameters, FLOPs, activations, or a structure summary.
- Treat MOT paths as project-specific and sequence-dependent; do not feed unrelated still images to a tracking workflow.
- If you only need to plan a command, stop at the helper output and do not run the underlying tool.

Files in this skill

  • SKILL.md2.1 KB
  • references/demo-workflows.md4.2 KB
  • references/tools-reference.md5.6 KB
  • references/troubleshooting.md2.2 KB
  • scripts/build_tool_command.py13.3 KB

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.