visual-anagrams
Routes visual-anagram generation, view selection, animation, and illusion metadata tasks.
- License 许可
- MIT
- 最近更新
- 2026-09-03
- Stars
- ★ 351
- Forks
- 27
01 前言 Front Matter
- name
- visual-anagrams
- license
- MIT
- metadata
- { "disco-role": "operating" }
- description
- Routes visual-anagram generation, view selection, animation, and illusion metadata tasks.
- disable-model-invocation
- true
02 资源文件 4
- 参考文档references/troubleshooting.md2.3 KB
- 参考文档references/views.md1.1 KB
- 参考文档references/workflows.md1.8 KB
- 脚本scripts/check_views.py1.9 KB
03 SKILL.md 正文
Visual Anagrams
Use this subskill when the user wants to generate a multi-view optical illusion, inspect the available views, or animate a finished illusion.
It covers the generate.py and animate.py branches in the Visual Anagrams subproject.
Include here
- Multi-view illusion generation.
- View selection and view-argument handling.
- Metadata save/load for later animation.
- Animation of an existing illusion image.
- Resolution and upsampling settings specific to the Visual Anagrams branch.
Exclude or route elsewhere
- Ordinary image generation: use
image-generation. - Training or finetuning: use
image-training. - Audio or music demos: use
audio-music. - ImageNet benchmark training: use
imagenet-training.
Read first
references/workflows.mdfor the generation and animation routes.references/views.mdfor the supported view names and their common arguments.references/troubleshooting.mdfor view-count, metadata, and dependency failures.scripts/check_views.pybefore generation if the view names or arguments are uncertain.
Fast routing hints
- If the user says
visual anagram,illusion,view, oranimate, use this subskill. - If the user only needs a standard image, stay in
image-generation.
04 同仓库更多 Skills
passnet-triton-opt
Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block/grid/warp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.
passnet-solve
END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.
passnet-skill
PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under ./pass_dir/, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.
passnet-pattern-fusion
Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()/replacement_args()/replacement_func() files, when a pass "failed to match", or when deciding how to fuse multiple kernels/ops.
passnet-orchestrate
Round/batch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep/revert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.
passnet-feedback
Fast, deterministic feedback for PassNet work: pre-flight pattern/match verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).