start-pixel-detective
Skill4 starsChanged 10 months agoMITView on GitHub
Start, stop, or check Pixel Detective services on this machine. Use when the user asks to start Pixel Detective, start backend only, start full stack, restart, stop, or check Docker, Qdrant, ML, UMAP, or frontend status, or to provide a specific Pixel Detective page URL.
In this repository71 other files it gives its agents.
- AGENTS.mdAGENTS.md
- algorithmic-artSkill
- architecture-diagram-creatorSkill
- artifacts-builderSkill
- brainstormingSkill
- brand-guidelinesSkill
- canvas-designSkill
- changelog-generatorSkill
- code-auditorSkill
- codebase-documenterSkill
- code-executionSkill
- code-refactorSkill
- code-transferSkill
- content-research-writerSkill
- conversation-analyzerSkill
- csv-data-summarizerSkill
- dashboard-creatorSkill
- dev-graph-exportsSkill
- dispatching-parallel-agentsSkill
- doc-coauthoringSkill
- docker-helperSkill
- docxSkill
- ensemble-solvingSkill
- executing-plansSkill
- feature-planningSkill
- file-operationsSkill
- file-organizerSkill
- finishing-a-development-branchSkill
- flowchart-creatorSkill
- frontend-designSkill
- git-pushingSkill
- image-enhancerSkill
- imagenSkill
- internal-commsSkill
- markitdownSkill
- mcp-builderSkill
- pdfSkill
- pixel-detective-exportsSkill
- pptxSkill
- project-bootstrapperSkill
- receiving-code-reviewSkill
- requesting-code-reviewSkill
- review-implementingSkill
- rituals-guideSkill
- systematic-debuggingSkill
- scientific-visualizationSkill
- skill-creatorSkill
- skills-summarizerSkill
- skills-updaterSkill
- slack-gif-creatorSkill
- sprint-perspectivesSkill
- sprint-summary-pdfSkill
- start-dev-graphSkill
- subagent-driven-developmentSkill
- systematic-debuggingSkill
- tailored-resume-generatorSkill
- technical-doc-creatorSkill
- template-skillSkill
- test-driven-developmentSkill
- test-fixingSkill
- The first 60 shown.
The file
We split this skill into the rules it gives an agent
10 rules. About 0 are common, 8 are distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈8
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 8
Use this skill to run Pixel Detective locally with the right subset of services.
- Distinctive: line 18
If missing, stop and ask for the correct working directory.
- Distinctive: line 26
stop: stop Docker services and list any remaining processes
- Distinctive: line 29
Check Docker CLI and Docker Compose:
- Distinctive: line 32
If Docker is installed but not running, say so and stop.
- Very specific: line 39
start ML service: uvicorn backend.ml_inference_fastapi_app.main:app --port 8001 --reload
- Distinctive: line 47
When launching uvicorn or npm, use new terminal windows so the main session stays responsive.
- Distinctive: line 65
Use existing startup scripts whenever possible to stay aligned with repo behavior.
- Very specific: line 67
If port 3000 is already in use, Next.js will choose 3001 or another port.
- Distinctive: line 68
If the ML or ingestion uvicorn process exits immediately, check for missing Python dependencies or a virtualenv not being activated.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Start Pixel Detective
Use this skill to run Pixel Detective locally with the right subset of services.
Quickstart
- Full stack:
powershell -ExecutionPolicy Bypass -File skills\\start-pixel-detective\\scripts\\start_pixel_detective.ps1 -Mode full - Backend only (+ optional GPU UMAP):
powershell -ExecutionPolicy Bypass -File skills\\start-pixel-detective\\scripts\\start_pixel_detective.ps1 -Mode backend -UseGpuUmap
Workflow
- Confirm repo root
- Require
frontend/package.jsonandstart_pixel_detective.ps1. - If missing, stop and ask for the correct working directory.
- Decide the requested mode
- full: all services plus frontend
- backend: Qdrant + ML + ingestion (optional UMAP)
- services: just Docker services (Qdrant and optional GPU UMAP)
- frontend: only Next.js dev server
- status: report what is running
- stop: stop Docker services and list any remaining processes
- Docker preflight
- Check Docker CLI and Docker Compose:
docker versiondocker compose version
- If Docker is installed but not running, say so and stop.
- Start sequence
- full: run
.\start_pixel_detective.ps1 - backend:
docker compose up -d qdrant_db- optional GPU UMAP:
docker compose -f backend/gpu_umap_service/docker-compose.dev.yml up -d --build - start ML service:
uvicorn backend.ml_inference_fastapi_app.main:app --port 8001 --reload - start ingestion:
uvicorn backend.ingestion_orchestration_fastapi_app.main:app --port 8002 --reload
- services:
docker compose up -d qdrant_db- optional GPU UMAP (same command as above)
- frontend:
cd frontendthennpm run dev
When launching uvicorn or npm, use new terminal windows so the main session stays responsive.
- Status checks
docker compose ps- The script prints a port readiness summary (6333/8001/8002/8003/3000).
- Report URLs:
- Stop sequence
docker compose down- The script attempts to close uvicorn and npm windows tied to this repo.
Notes
- The GPU UMAP service can fail without blocking the rest. Treat it as optional.
- Use existing startup scripts whenever possible to stay aligned with repo behavior.
- Qdrant is the only required Docker service for Pixel Detective; ML and ingestion run on the host.
- If port 3000 is already in use, Next.js will choose 3001 or another port; report the actual port from the terminal output.
- If the ML or ingestion uvicorn process exits immediately, check for missing Python dependencies or a virtualenv not being activated.
Script
- Run:
powershell -ExecutionPolicy Bypass -File skills\\start-pixel-detective\\scripts\\start_pixel_detective.ps1 -Mode full - Backend only:
-Mode backend -UseGpuUmap - Frontend + open page:
-Mode frontend -Open -Page /search - Note:
-Openis required to open a browser;-Pagealone will not open anything.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

