media-tools
Use when generating visual artifacts — ASCII art, halftone, dithering, matrix-rain, VHS, pixel-sort, voronoi, and other image/video FX — from images, GIFs, videos, GLB models, or webcam input. Local, offline-first.
- License 许可
- MIT
- 最近更新
- 2026-11-15
- Stars
- ★ 0
- Forks
- 1
01 前言 Front Matter
- name
- media-tools
- triggers
- [ "user" ]
- description
- Use when generating visual artifacts — ASCII art, halftone, dithering, matrix-rain, VHS, pixel-sort, voronoi, and other image/video FX — from images, GIFs, videos, GLB models, or webcam input. Local, offline-first.
- argument-hint
- What visual artifact should I generate?
03 SKILL.md 正文
Media Tools
Local visual-FX generators (grainrad.com pipeline parity). Turns inputs — images, GIF/MP4/WebM, GLB 3D models, webcam frames — into stylized PNG/JPEG/ SVG/GIF/MP4 artifacts. All offline; JSON on stdout.
Where it is
- Bundle source:
extensions/media-tools/ - Installed:
%APPDATA%\devin\extensions\media-tools\(Windows) or~/.config/devin/extensions/media-tools/(POSIX) - Deps:
pillow,numpy,opencv-python— pinned inrequirements.txt, install on demand:python -m pip install -r requirements.txt(no installer venv).cv2is only needed for webcam/video input.
Tools
grainrad.py— unified FX CLI.--input(file |webcam[:N]),--effect,--param k=v(repeatable; prefixadjust./process./bloom./grain./chromatic./scanlines./vignette.for pipeline stages),--preset,--frames,--fps,--format png|jpeg|svg|txt|threejs.ascii_mancer.py— dense+block ASCII blend → PNG.--cell-detail,--cell-block.glb_input.py— glTF-Binary loader/renderer (orthographic, flat shading); used via--input model.glbin grainrad.
Effects (--list-effects): ascii, blockify, contour, crosshatch,
dithering, dots, edgeDetection, halftone, matrixRain, noiseField,
pixelSort, threshold, vhs, voronoi, waveLines.
Flow
# discover effects/presets
python grainrad.py --list-effects
python grainrad.py --list-presets
# still image → stylized PNG
python grainrad.py --input photo.png --effect halftone --output out.png
# animated input → GIF (10fps default)
python grainrad.py --input clip.mp4 --effect vhs --output out.gif --fps 12
# webcam capture → effect
python grainrad.py --input webcam:0 --effect ascii --output selfie.png
Every call prints {"ok": true, "path": ...} — verify ok before reporting
done. --self-test runs synthetic offline asserts per script.
External catalog — tooooools.app
If the user asks for a visual effect or media process NOT covered by the local
generators above, MANDATORY: read
extensions/media-tools/knowledge_bases/creative_tooooools.json first — a
curated catalog (16 browser tools: stippling, dithering, CRT, cellular
automata, displacement, ASCII, animate…) with name, URL, and description per
tool. Use it to suggest the right external tool or as reference to implement a
local equivalent. Refresh with python update_tooooools_db.py.
Craft suite dispatch (structured media ops)
For document-grade operations (timeline edits, PSD layers, PDF,
sheets, slides, CAD, audio) prefer the storytold craft engines over
reimplementing: every craft app is agent-native — <app>-cli commands,
<app>-cli run, <app>-cli mcp (MCP server), some add --control
JSON-lines. All Apache-2.0; spark is MIT for gsplat preview.
- Bridge:
extensions/craft-bridge/bridge.py—doctor,commands <app>,run <app> <command_id> [params]normalized across the suite. - Manifests:
scene_manifest.pycompiles ourscene_manifest.json(schema v1) into engine command plans; unknown ops pass through markedunverified— confirm ids viabridge.py commands <app>first. - Coverage map:
extensions/craft-bridge/ops_matrix.md. - gsplat preview:
extensions/craft-bridge/spark-view/(static three.js- spark page, opens .PLY/.SPZ).
- MCP channels: register lazily per
mcp-lazy-enablement— enable the one app server the task needs, never the whole suite.
Boundary
- This skill is for visual artifact generation (stylized media).
- UI/UX polish of interfaces →
impeccable. Diagram artifacts (C4/topology) →architecture-diagrams. Launch videos from project sites →brag.
04 同仓库更多 Skills
montador-de-ideias
Use when assembling, expanding, or elevating a raw project idea or initial prompt into an ultra-detailed, first-principles dossier and compiled master prompt with journalistic OSINT market research, Tech Lead and Staff Architect justifications, atomic and modular hardening/optimization, and high-impact contact surfaces.
git-github-adaptive-engineering
Use when any task touches git history, GitHub issues/PRs/releases/CI, repo governance, or branch and merge decisions. Runs the Fase Zero situational-awareness battery, picks one of five delivery flows via the adaptive decision engine, and enforces commit/PR information-density standards. Orchestrates the gh, git-workflows, github-actions, pr, and debug/finish skills.
adaptive-agenda-calendar-routine-engine
Use when planning or executing work across hours, days, weeks, or sprints. Turns the calendar into a strict time-blocked agenda (HH:MM-HH:MM blocks carrying a routine profile, expected inputs, and a handoff output artifact), runs the 5-stage cognitive flow (Radar and Hunt -> Triage plus Steve Jobs conception -> workbench manifest -> >=32-cycle loop -> Close plus Open-Minded Scrum), and unfolds macro calendar milestones into micro agenda blocks through extensions/gcal/wrapper.py.
demo
Use when demonstrating a skill in a fixture project.
duplicate-skill
Use when testing duplicate skill name detection.
duplicate-skill
Use when testing duplicate skill name detection.