AI Skills
Installable SKILL.md packages that extend AI coding agents with reliable workflows and domain-specific judgment — ranked by installs, sourced from skills.sh.
setup-matt-pocock-skills
One-time repo setup for Matt Pocock's engineering skills — configures issue tracker, triage labels, and domain doc layout.
codex-pet
Generate a custom OpenAI Codex Pet spritesheet from one image — no Codex Pro or OPENAI_API_KEY required.
lark-markdown
View, create, upload, edit, patch, and diff Markdown files stored in Lark (Feishu) Drive
azure-upgrade
Assess and upgrade Azure workloads between plans, tiers, SKUs, or modernize Azure SDK dependencies in source code.
triage
Triage issues and PRs through a state machine: categorise, verify, grill, and produce agent-ready briefs.
ai-video-generation
Generate AI videos via CLI — smart model routing across HappyHorse, Kling, Seedance, Veo, Wan, Hailuo, and Dreamina.
ai-image-generation
Generate & edit images with 11+ AI models via one CLI command — smart model routing for text-to-image and image-to-image.
runcomfy-cli
One CLI to run any RunComfy model — image, video, edit, lip-sync, face-swap, LoRA training, and more.
ai-avatar-video
Generate AI avatar and talking-head videos with lip-sync via RunComfy CLI — OmniHuman, HappyHorse, Seedance, Wan 2-7
face-swap
Swap faces or characters into images and video using the RunComfy CLI — routes across 5 models by intent.
lark-vc-agent
Real-time meeting participation and event monitoring for Feishu video conferences.
video-inpainting
Remove objects, watermarks, and unwanted regions across video frames using prompt-driven AI editing.
lark-approval
Query, process, and initiate Lark approval tasks, instances, and definitions.
image-inpainting
Mask-driven image inpainting for object removal, watermark cleanup, and region replacement via RunComfy CLI.
controlnet-pose
Pose-conditioned image and video generation via RunComfy CLI—transfer motion, control character stance, or apply depth/canny conditioning.
lipsync
Sync audio to a face's mouth across multiple AI models via RunComfy CLI.
video-extend
Extend existing video clips with Google Veo 3-1 via RunComfy CLI.
image-outpainting
Extend image canvas, change aspect ratio, and fill in missing areas while preserving original content.
elevenlabs-music-generation
Generate studio-quality songs and instrumentals from text descriptions via ElevenLabs Music on RunComfy.
video-outpainting
Extend video canvas spatially—swap aspect ratio, uncrop, add environment while preserving central action.
relight
Change lighting direction, color temperature, and mood on still images via RunComfy's Qwen Edit Relight LoRA.
supabase-postgres-best-practices
Postgres performance optimization and best practices guide from Supabase for query, schema, and configuration tuning.
github-actions-docs
Ground GitHub Actions questions in official documentation with exact links and docs-backed YAML guidance.
azure-enterprise-infra-planner
Architect and provision enterprise Azure infrastructure from workload descriptions with Bicep or Terraform.
ai-music
Generate AI music via RunComfy CLI—route to ElevenLabs premium vocals or cheap open-weights ACE Step, plus audio editing.
ace-step
Generate, inpaint, and outpaint music with ACE Step on RunComfy—tag-driven composition at $0.0002–0.0003/s.
openclaw-secure-linux-cloud
Secure self-hosted OpenClaw on Linux cloud servers with conservative defaults: loopback binding, SSH tunneling, token auth, and minimal tool permissions.
secure-linux-web-hosting
Set up and harden a Linux web server for self-hosting with DNS, SSH, Nginx, HTTPS, and optional tuning.
skills-cli
Package manager for discovering, installing, and managing installable Agent Skills via CLI.
running-claude-code-via-litellm-copilot
Route Claude Code through a local LiteLLM proxy to GitHub Copilot, reducing Anthropic API spend.