jack-tar-mlx

Full presentation pipeline — brand, style, narrative, images, SmartArt, assembly, QA + academic-figure rendering (optional paperbanana CLI; local-first Ollama tier + local_only zero-spend mode, #119) + /iterate-slide single-slide critique-driven refinement + full_bleed image-is-the-slide strategy + creative vision renderer (#105) + creative_vision GA: per-slide cost surface, Creative Sprint phase, per-iteration operator gate, deck-level creative anchors (#113) + annotate-figure v2.1: composed-strategy annotation zone + native full-slide headline opt-in (#142 v2.1) + blank-zone annotation variant: reserve an empty image region for label placement, verified + fallback-guaranteed (#142) + edit tier: $0 local mflux targeted edits as a fourth refinement channel in iterate-slide and the creative_vision cascade (#143)

13 skills · 9 agents · hooks

Cloud AI image generation — OpenAI, Google, FAL.ai, Recraft with smart provider routing

9 skills · 1 agent

Local AI image generation via Ollama — images, icons, patterns, diagrams, and quick presentations

6 skills · 1 agent

Wraps document-skills:pptx with narrative pre-brief (/bridge-brief) and post-hoc visual enrichment (/enrich-deck) — combine /pptx structural strength with jack-tar visual capabilities (AI backgrounds, element images, editable SmartArt, native charts)

3 skills · 2 agents

Editable PowerPoint SmartArt graphics — 29 layouts, speakers can modify after delivery

4 skills

Data visualisation and custom graphics — SVG, Mermaid, Vega-Lite, matplotlib charts

3 skills

OpenAI GPT Image 2/2.5/2.6 skill — Image 2.6 Flare by default, on-demand prompt references, and a packaged CLI for generation and editing.

2 skills

Skills for MLX-VLM: command-line and server inference, convert/quantize, adding new models, benchmarking, contributing, cached-model listing, and reproducible GitHub issues.

8 skills

Generate and edit images with OpenAI GPT Image 2 and Gemini 3.1 Flash Image Preview.

1 skill · 4 commands

Turn Claude Code into a visual creative assistant — AI image generation with creative workflow orchestration, prompt engineering, parallel multi-direction output, and 1,300+ curated inspiration library

2 commands · 3 agents · 1 MCP server

Operator-grade Ideogram skill pack with 24 current V4, async, security, storage, governance, and migration workflows

24 skills

Image/video generation, complete brand systems, narrated explainers, YouTube thumbnails, browser games, Marketing Studio, video engagement analysis, and Soul Character training via the Higgsfield CLI.

9 skills