alexzh3/openhumanoid/perception-stack

Claude Code and agent skill by alexzh3/openhumanoid

Trust tier 3
  • community
Source check pending

Checking that the SKILL.md file is still in the repository.

Use the camera-driven perception stack for scene understanding, 3D object grounding, and face recognition.

Install

npx vskill@latest install alexzh3/openhumanoid/perception-stack

SKILL.md

Perception Stack

The perception stack lives at http://localhost:8787 unless CAPABILITY_SERVER_URL says otherwise.

Understand the current scene

bash
curl -s -X POST http://localhost:8787/perception/scene \
  -H 'Content-Type: application/json' \
  -d '{"camera_name":"zed-mini"}'

Find a specific object and return its 3D pose

bash
curl -s -X POST http://localhost:8787/perception/object_pose \
  -H 'Content-Type: application/json' \
  -d '{"label":"apple","color":"green"}'

Turn a grounded object pose into a grasp candidate

bash
curl -s -X POST http://localhost:8787/perception/grasp_pose \
  -H 'Content-Type: application/json' \
  -d '{"object_id":"apple-green-01","pose":{"x":2.78,"y":1.08,"z":0.79,"frame":"map"}}'

Enroll a face

bash
curl -s -X POST http://localhost:8787/perception/face/enroll \
  -H 'Content-Type: application/json' \
  -d '{"person_id":"alice","display_name":"Alice"}'

Recognize known faces

bash
curl -s -X POST http://localhost:8787/perception/face/recognize \
  -H 'Content-Type: application/json' \
  -d '{"min_confidence":0.8}'

Usage notes

  • Use perception before navigation when the task depends on finding an object or support surface.
  • Use perception again after navigation to refine the close-range pose before manipulation.
  • Use perception/grasp_pose when the downstream manipulation step needs an explicit wrist target rather than just an object ID.
  • Face recognition should be treated as identity verification with a confidence threshold, not as a free-form VLM guess.

Security

  • Security scanPassed, 90 of 100Pattern checks on the skill's files.
  • Model reviewPassed

See the full security report

Quality evaluation

  • Task evaluationNot runNo evaluation yet. Run one in Skill Studio.

See evaluation results

Source

Popularity

7-day trend score
5

Not refreshed yet.

Works with

Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.

  • Claude Code
  • Cursor
  • GitHub Copilot
  • Windsurf
  • Codex
  • Gemini CLI
  • Cline
  • Amp
  • Roo Code
  • Goose
  • Continue
  • Kilo Code
  • Junie
  • Trae
  • OpenHands
  • Qwen Code
  • AI Agent
  • Tabnine Chat
  • Sourcegraph Cody
  • Amazon Q
  • Firebase Genkit
  • Sweep
  • Devin
  • Phind
  • Mentat
  • Claude Desktop
  • Aider
  • GPT Pilot
  • Plandex
  • bolt.new
  • v0
  • Replit Agent
  • Anse
  • TypingMind
  • BetterChatGPT
  • LibreChat
  • Warp AI
  • Pieces
  • Fig AI
Embed badge
vskill VERIFIED badge
[![vskill VERIFIED](https://verified-skill.com/api/v1/skills/alexzh3/openhumanoid/perception-stack/badge)](https://verified-skill.com/skills/alexzh3/openhumanoid/perception-stack)

Something wrong with this skill or its page? Tell us.