MCP App StoreMCP App Store
from sprite-gen132
Create clean 2D game sprite sheets and animated atlases from image-generation rows: chroma-key cleanup, connected-component frame extraction, atlas composition,
Sprite Gen provides a full pipeline for turning image-generation outputs into runtime-ready sprite sheets and manifests. It handles run preparation, chroma-key alpha cleanup, connected-component extraction, frame composing, QA previews, and an interactive curation webview for human selection and transforms.
Use this skill when you need automated sprite-sheet generation from generative image outputs, to curate image candidate sets side-by-side, or to rebuild and edit existing atlases. It's aimed at game developers and artists producing 2D animation assets.
Works well with Codex and Claude Code environments that can run Python tooling and launch local webviews; integrates with image-generation backends (kuma:image-gen or similar).
Sprite-gen is a highly detailed, opinionated skill for generating 2D game sprite sheets and animated atlases via a multi-stage pipeline (prepare → generate → extract → curate → compose). The SKILL.md is exceptionally thorough with blocking gates, SSoT contracts, Korean/English bilingual instructions, and extensive doc topology. Scripts couldn't run because they require a dedicated venv with sprite_gen package installed — 22/24 are thin wrappers, 1 is a mutation battery test, 1 is a measurement tool. No security concerns found: no hardcoded creds, no curl|bash, no exfiltration, no telemetry, no destructive commands. The skill explicitly mandates using its own venv interpreter and refuses to fall back to system python3. The serve_curation webview includes path traversal protection.
sprite_gen (requires venv with Pillow, NumPy installed via pip install -e .)No security issues detected. Skill is well-structured with clear contracts, gating mechanisms, and documentation topology. The bilingual (KR/EN) SKILL.md is thorough but very long — could benefit from progressive disclosure via the referenced docs/. Architecture follows skill spec reasonably well with scripts/ directory and docs/ references. Usefulness is moderate: niche audience (game devs making pixel art sprites), heavy setup requirements (venv, image gen providers), but genuinely solves a real problem for that audience.
Retriever (Membrane CLI)
Integrate Retriever search and discovery via the Membrane CLI — manage connections and run queries across Retriever data sources.
Seq Wrangler
Sequence data QC, trimming, alignment, and BAM processing pipeline helper — wraps FastQC, fastp/Trimmomatic, BWA/Bowtie2/Minimap2, and SAMtools for automated re
Cost Tracker
Monitor agent session costs, set budget alerts, and get actionable token-spend optimizations to keep multi-session workflows within budget.
Retriever (Membrane) Integration
Integrate with Retriever via the Membrane CLI to search, query, and proxy Retriever data (documents, snippets, queries) with built-in auth and pagination.
OpenTestAI
Automated, high-confidence AI testing: bug detection, persona feedback, and prioritized test-case generation using many specialized tester profiles.
Codex Collab
Bridge Claude and Codex to run tasks, perform code reviews, and run background jobs for research or implementation workflows.
Manim Idea to Export
Turn plain-language concepts into production-ready ManimCE scene blueprints, runnable code, preview renders, and final export commands for video assets.
PPT as Code
Creator-first workflow to plan and build HTML-first presentations with staged modes (quick/basic/advanced), optional PPTX export, and clear QA and image plannin