Skip to content
Back to search

agent-meme-forge

Codex skill that turns one image or text concept into a WeChat-ready Chinese sticker pack people will actually send. Sendability-gated, 4-keypose → 16-frame, no model-drawn text.

Safety scan

A static text scan, not a guarantee. Select the badge to see what was scanned and every finding.

Checked today

Check it deeper on your machine

Its scripts were only partly scanned.

As a Claude Code plugin, install it and then ask Claude Code:

/plugin marketplace add keithmackay/mackayi/plugin install scanskillsafety@mackayiDo a deep scan of https://github.com/lisamsung/agent-meme-forge with scanskillsafety

Or from a terminal, clone it and run it:

git clone https://github.com/keithmackay/scanskillsafety-skill ~/.claude/skills/scanskillsafetynode ~/.claude/skills/scanskillsafety/cli.cjs --deep 'https://github.com/lisamsung/agent-meme-forge'

It checks the exact commit: every file, its packages against OSV (api.osv.dev), how it compares with this page's scan (findsafeskills.com receives the repository's address), and, if you agree, Claude reading every file with your own Claude subscription. It never runs the code.

Badge for your README
findsafeskills safety rating

Markdown

[![findsafeskills safety rating](/api/listings/lisamsung--agent-meme-forge/badge.svg)](/listings/lisamsung--agent-meme-forge)

HTML

<a href="/listings/lisamsung--agent-meme-forge"><img src="/api/listings/lisamsung--agent-meme-forge/badge.svg" alt="findsafeskills safety rating" /></a>

Install

i

Install instructions from the README

As written by the repo's author. findsafeskills has not verified these steps.

# 1. clone and install
git clone https://github.com/lisamsung/agent-meme-forge.git
cd agent-meme-forge
python3 -m venv .venv && . .venv/bin/activate
pip install -e '.[dev]'

# 2. plan a pack interactively
python skills/generate-meme-gif-pack/scripts/meme_pack.py plan-wizard

# 3. after you've generated and accepted the keypose sheets,
#    build the WeChat upload package
python skills/generate-meme-gif-pack/scripts/meme_pack.py build-pack \
  --source-dir output/raw-frames/MyPack \
  --output-dir output/my-pack \
  --persona 科研打工人 \
  --style clean-sticker \
  --pack-size 24 --mode wechat \
  --pack-name 我的表情包 \
  --source-mode keyposes --keypose-layout 2x2 --render-frame-count 16 \
  --quality-mode submission --strict-qc --strict-continuity-qc

Open `output/my-pack/preview.html` to inspect the full pack, then upload `wechat-submit/` to the WeChat Sticker Open Platform.

> **Tip — first 3 first.** Don't generate 24 keypose sheets in one go. Run `build-preview --preview-count 3` after the first three sheets pass `qc-sheet`. If they aren't sendable yet, fix captions or motion templates before scaling out.

What was scanned

The files the safety scan read, as written by the repository's author and shown as plain text. Lines with a finding are marked.

README.md · README · 183 lines

View this file on GitHub (read at the repository's current version: the scan did not record which commit it read)

  1. # agent-meme-forge
  2. > **没人用的表情包就是垃圾。** A Codex skill that turns one reference image — or a text concept — into a WeChat-ready animated Chinese sticker pack that people will actually send.
  3. [![License: MIT](https://img.shields.io/badge/License-MIT-black.svg)](LICENSE)
  4. [![Python 3.10+](https://img.shields.io/badge/python-3.10%2B-blue.svg)](pyproject.toml)
  5. [![CI](https://github.com/lisamsung/agent-meme-forge/actions/workflows/ci.yml/badge.svg)](https://github.com/lisamsung/agent-meme-forge/actions/workflows/ci.yml)
  6. [![GitHub release](https://img.shields.io/github/v/release/lisamsung/agent-meme-forge)](https://github.com/lisamsung/agent-meme-forge/releases/latest)
  7. [![Codex Skill](https://img.shields.io/badge/codex-skill-ff4f64.svg)](skills/generate-meme-gif-pack/SKILL.md)
  8. [![中文 README](https://img.shields.io/badge/README-中文-green.svg)](README.zh-CN.md)
  9. [![Live demo](https://img.shields.io/badge/demo-GitHub%20Pages-2563eb.svg)](https://lisamsung.github.io/agent-meme-forge/)
  10. <p align="center">
  11. <img src="docs/assets/ai-mascot-shoudao-lixian.gif" alt="HR已读" width="180" />
  12. <img src="docs/assets/ai-mascot-jiazaizhong.gif" alt="录用快来" width="180" />
  13. <img src="docs/assets/ai-mascot-wenxianshan.gif" alt="先睡为敬" width="180" />
  14. </p>
  15. <p align="center"><sub>Three stickers from a 24-pack for <code>法律硕士毕业求职人</code>: HR-read silence, summoning an offer, and giving up to sleep. All produced from clean 4-keypose sheets and rendered locally to 16-frame 240×240 GIFs.</sub></p>
  16. ---
  17. ## Why this exists
  18. Most "AI sticker pack" tools optimize for *cute*. Cute doesn't get sent in WeChat. **Sendable** does.
  19. Every sticker this skill plans must clear a `sendability_gate`:
  20. 1. **Reuse trigger** — a real chat moment where someone would actually send it.
  21. 2. **Emotional value** — relief, sarcasm, agreement, panic, comfort, delay.
  22. 3. **Creative hook** — an idea, not just a caption pasted on a face.
  23. 4. **Visual gag** — readable at 240×240, on a phone, in one second.
  24. If a sticker is only decorative, only cute, only "vibe" — the skill is told to rewrite it before generation, not after.
  25. ## What it builds
  26. ```
  27. your reference image ┐
  28. or text concept ├──▶ plan ──▶ 4 keyposes ──▶ 16-frame GIF ──▶ WeChat upload pack
  29. + persona + style ┘ (JSON) (one image_gen (local render, (24 items,
  30. call per sticker) Chinese captions) manifest, QC report)
  31. ```
  32. A finished build looks like this:
  33. ```
  34. output/my-pack/
  35. ├── named-gifs/ # 收到离线.gif, 加载中.gif … (for sharing)
  36. ├── wechat-submit/
  37. │ ├── main/01.gif … 24.gif # numbered upload sequence
  38. │ ├── thumbs/01.png … 24.png # 120×120 thumbnails
  39. │ ├── cover.png icon.png banner.png
  40. │ └── reward-guide.png # optional, when 接受赞赏 is enabled
  41. ├── preview.html # browser preview of the whole pack
  42. ├── manifest.json / .csv # full machine-readable build manifest
  43. └── qc_report.json # per-sticker QC + continuity gates
  44. ```
  45. Under the hood:
  46. - **Plan** — character card + 24 sendable meme entries + per-sticker no-text `image_gen` prompts.
  47. - **Render** — 4 keyposes → 16 deterministic frames, using motion templates (`soul_offline`, `loading_loop`, `pretend_understand`, …) with local effects (soul puffs, loading dots, sweat drops, awkward lines). The image model never has to draw 16 coherent frames.
  48. - **QC** — fake-checkerboard detection, edge-touch, bbox drift, prop position/area jumps, one-frame prop flicker, face/head-shape drift, loop closure, low-motion-energy guard. Strict mode for WeChat submission.
  49. - **Pack** — 16 or 24 GIFs at 240×240 < 500 KB, thumbnails, cover, icon, banner, dual filename schemes (numbered for upload, Chinese for sharing), CSV + JSON manifest.
  50. - **Submit (optional)** — Playwright + headed Microsoft Edge runbook for the WeChat Sticker Open Platform, with QR-login, CAPTCHA, real-name, and rejection-flow handling.
  51. ## Quickstart
  52. ```bash
  53. # 1. clone and install
  54. git clone https://github.com/lisamsung/agent-meme-forge.git
  55. cd agent-meme-forge
  56. python3 -m venv .venv && . .venv/bin/activate
  57. pip install -e '.[dev]'
  58. # 2. plan a pack interactively
  59. python skills/generate-meme-gif-pack/scripts/meme_pack.py plan-wizard
  60. # 3. after you've generated and accepted the keypose sheets,
  61. # build the WeChat upload package
  62. python skills/generate-meme-gif-pack/scripts/meme_pack.py build-pack \
  63. --source-dir output/raw-frames/MyPack \
  64. --output-dir output/my-pack \
  65. --persona 科研打工人 \
  66. --style clean-sticker \
  67. --pack-size 24 --mode wechat \
  68. --pack-name 我的表情包 \
  69. --source-mode keyposes --keypose-layout 2x2 --render-frame-count 16 \
  70. --quality-mode submission --strict-qc --strict-continuity-qc
  71. ```
  72. Open `output/my-pack/preview.html` to inspect the full pack, then upload `wechat-submit/` to the WeChat Sticker Open Platform.
  73. > **Tip — first 3 first.** Don't generate 24 keypose sheets in one go. Run `build-preview --preview-count 3` after the first three sheets pass `qc-sheet`. If they aren't sendable yet, fix captions or motion templates before scaling out.
  74. ## Install as a Codex skill
  75. ```bash
  76. mkdir -p ~/.codex-switcher/skills
  77. cp -R skills/generate-meme-gif-pack ~/.codex-switcher/skills/
  78. # restart your Codex session
  79. ```
  80. Then ask Codex something like *"做一个科研打工人的微信表情包"* — it will run intake, plan the pack, and walk you through keypose generation and QC across turns.
  81. ## How it works
  82. The skill is built around two product decisions that are unusual for a sticker tool:
  83. **1 · The image model only draws 4 stable keyposes per sticker.**
  84. Asking a diffusion model to freely draw 16 coherent frames produces jumpy, unrelated frames most of the time. Instead, this skill asks for a clean `2×2` sheet of *start / anticipation / peak gag / loop return*, and a deterministic local renderer interpolates the 16 frames, holds, anticipation, rebound, and loop closure. Motion templates (`soul_offline`, `loading_loop`, `pretend_understand`, …) layer on the comic effects locally — soul puffs, loading dots, sweat lines — so the model can't randomly invent or skip them.
  85. **2 · No Chinese text is ever drawn by the model.**
  86. Captions are typeset by the local processor with consistent fonts, sizing, and clipping rules, so 24 stickers actually look like one pack. Visual prompts hard-ban text, captions, speech bubbles, UI, and logos.
  87. The result: more reliable identity, fewer reroll cycles, predictable WeChat-spec output.
  88. **Smoothest path — dense real frames (recommended).** Decision #1 keeps keyposes as the safe default that runs on any provider. When the image model is strong enough to draw ~8 genuinely different, consistent frames in a single sheet, the `dense_frames` source mode asks for exactly that — and a local pass normalizes every frame to a uniform size and head position, times them, and engineers the loop. It's the recommended path for maximum smoothness (expect a few stickers per pack to need regenerating). See **[dense real frames](skills/generate-meme-gif-pack/references/dense-frames.md)**.
  89. ## WeChat spec defaults
  90. | Asset | Spec |
  91. |---|---|
  92. | Main GIF | `240×240`, looped, < 500 KB (target < 480 KB) |
  93. | Thumbnail PNG | `120×120`, < 50 KB |
  94. | Album icon | `50×50` transparent, < 30 KB |
  95. | Cover | `240×240` transparent, < 80 KB |
  96. | Detail banner | `750×400`, no text, < 80 KB |
  97. | Pack size | **16 or 24** (this skill defaults to 24) |
  98. | Optional reward guide | `750×560`, when `接受赞赏` is enabled |
  99. | Optional reward thanks | `750×750`, when `接受赞赏` is enabled |
  100. Always re-verify against the WeChat Sticker Open Platform docs before commercial submission — official rules change.
  101. ## Project layout
  102. ```
  103. skills/generate-meme-gif-pack/
  104. ├── SKILL.md # agent rules and workflow
  105. ├── scripts/meme_pack.py # plan → accept → qc → preview → build pipeline
  106. └── references/
  107. ├── personas.md styles.md meme-library.md
  108. ├── prompt-rules.md # raw image prompting + sendability gate
  109. ├── wechat-spec.md # output constraints
  110. └── wechat-platform-upload.md # optional Playwright submission runbook
  111. docs/ # GitHub Pages product site + sample assets
  112. tests/ # pytest suite (run with `pytest -q`)
  113. ```
  114. ## Docs
  115. - [Live product page](https://lisamsung.github.io/agent-meme-forge/)
  116. - [Skill rules and workflow](skills/generate-meme-gif-pack/SKILL.md)
  117. - [Dense real frames — recommended for smoothness](skills/generate-meme-gif-pack/references/dense-frames.md)
  118. - [Example test report](docs/example-test-report.md)
  119. - [WeChat public-account draft (中文)](docs/wechat-public-account-draft.md)
  120. - [中文 README](README.zh-CN.md)
  121. - [Changelog](CHANGELOG.md)
  122. - [Release process](docs/RELEASING.md)
  123. ## Status and contributing
  124. - Test suite: `pytest -q` (covers WeChat size constraints, keypose planning, motion-template rendering, continuity QC, prop-flicker / face-drift gates, sheet splitting, caption layout, and full submission-pack structure).
  125. - Verified end-to-end on four sample packs: 24 × `科研打工人`, 16 × `码农`, 18 × `都市丽人` (self-use), and a 24-pack of an original AI mascot in `2×4` strict-QC mode.
  126. - WeChat platform rules can change at any time — please re-check the official help center before commercial submission.
  127. Issues and pull requests are welcome, especially for new motion templates,
  128. persona libraries, local effects, and reproducible QC improvements. Read
  129. [CONTRIBUTING.md](CONTRIBUTING.md) before proposing a substantial change. Usage
  130. questions belong in [GitHub Discussions](https://github.com/lisamsung/agent-meme-forge/discussions),
  131. and vulnerabilities should follow the private process in [SECURITY.md](SECURITY.md).
  132. ## Image rights & content review
  133. If your reference image is a real person, you must own the rights or have explicit permission. The default policy is "stylized but recognizable", **not** photo-realistic face swap. For public publishing, avoid politics, hate, sexual content, slurs, doxxing, medical claims, and harassment — WeChat review will reject these and so will this skill.
  134. ## License
  135. Project code is [MIT licensed](LICENSE). The bundled ZCOOL KuaiLe font is
  136. distributed under the SIL Open Font License 1.1; see
  137. [THIRD_PARTY_NOTICES.md](THIRD_PARTY_NOTICES.md).

Ratings and comments

Sign in to rate this skill.

Sign in

Free to use. If it’s helped, pass it on.