微信公众号贴图发布器

v1.0.3

Publish WeChat Official Account image-first sticker posts (图片消息 / 发贴图) into the draft box via the official API. Use when the user wants to create公众号“发贴图”草稿、图...

0· 128·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for hzlawliet/wechat-sticker-publisher.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "微信公众号贴图发布器" (hzlawliet/wechat-sticker-publisher) from ClawHub.
Skill page: https://clawhub.ai/hzlawliet/wechat-sticker-publisher
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install wechat-sticker-publisher

ClawHub CLI

Package manager switcher

npx clawhub@latest install wechat-sticker-publisher
Security Scan
Capability signals
Requires OAuth tokenRequires sensitive credentials
These labels describe what authority the skill may exercise. They are separate from suspicious or malicious moderation verdicts.
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
medium confidence
Purpose & Capability
Name/description match the implementation: the script uploads local images as permanent materials and creates a 'newspic' draft via api.weixin.qq.com. The requested actions are appropriate for publishing image-first drafts.
Instruction Scope
SKILL.md and the script read local absolute image files, load credentials from environment or a local wechat.env file, call WeChat endpoints, and write JSON output files under outputs/. These behaviors are consistent with the stated purpose but you should expect local image data to be transmitted to WeChat and API responses (media IDs, draft response) to be stored on disk.
Install Mechanism
No install spec or external downloads; this is an instruction-only skill with a small included Python script that depends only on the widely used 'requests' package (requirements.txt). No remote installers or unusual URLs are used.
!
Credentials
The registry metadata lists no required env vars, and SKILL.md marks WECHAT_APP_ID and WECHAT_APP_SECRET as optional, but the script calls require_creds() and will abort if those two variables are not present. The credentials requested are appropriate for the WeChat API, but the mismatch between declared requirements and the code's mandatory credentials is misleading and could cause confusion or accidental credential exposure if not handled carefully.
Persistence & Privilege
Skill is not always: true and doesn't request system-wide changes. It will create an outputs/ directory in its repo root and write API responses there; that per-skill persistence is expected and scoped to the skill.
Assessment
This skill appears to do what it says: it uploads local images to WeChat and creates drafts using the official API. Before installing or running it, consider the following: - You must provide WECHAT_APP_ID and WECHAT_APP_SECRET (the script requires them although the metadata suggests they are optional). Provide these via environment variables or a local wechat.env file; do NOT commit that file to source control. - The script will upload the actual image files to api.weixin.qq.com; ensure you intend to transfer those images. - The script writes JSON debug/output files to outputs/ that include upload responses and draft results (media IDs and API response bodies). Inspect those files before sharing—they may contain identifiers you don't want public. - No third-party or obfuscated code is downloaded, but run this in a controlled environment and review outputs if you will use real credentials. If you need higher assurance, ask the author to fix the metadata to clearly mark WECHAT_APP_ID/WECHAT_APP_SECRET as required and to document exactly what is written to outputs/.

Like a lobster shell, security has layers — review code before you run it.

Runtime requirements

🖼️ Clawdis
latestvk977qfq7dg7atddk22qjmj3521856r6a
128downloads
0stars
4versions
Updated 1w ago
v1.0.3
MIT-0

WeChat Sticker Publisher

Create draft-only WeChat Official Account sticker/image-message posts using the official API.

Workflow

  1. Use scripts/publish_sticker.py.
  2. Provide one or more absolute image paths.
  3. Provide a title and short text.
  4. The script uploads permanent image materials.
  5. The script creates a newspic draft in the WeChat draft box.
  6. Do not auto-publish; final publish is manual in the WeChat backend.

Rules

  • Keep this skill limited to draft creation.
  • Do not modify the existing daily article publishing workflow unless explicitly requested.
  • Require absolute local image paths.
  • Support up to 20 images; the first image becomes the cover image.
  • Send JSON request bodies explicitly as UTF-8 (application/json; charset=utf-8).
    • This is important: using a different request-body path can lead to garbled Chinese text.

Credentials

The script reads credentials from environment variables:

  • WECHAT_APP_ID
  • WECHAT_APP_SECRET

It can also load a local wechat.env file placed in the skill root. Use wechat.env.example as the starting template.

If packaging/publishing this skill publicly, do not include any real secrets or local credential files.

Primary command

python3 scripts/publish_sticker.py \
  --image /abs/path/to/image.jpg \
  --title "标题 / Title" \
  --text "配文 / Caption"

Multi-image example

python3 scripts/publish_sticker.py \
  --image /abs/1.jpg \
  --image /abs/2.jpg \
  --title "多图贴图测试" \
  --text "这是一个多图图片消息草稿。"

Outputs

Each run writes a JSON record to outputs/ containing:

  • uploaded image material responses
  • draft request body
  • draft creation response

Use those files for debugging and verification.

Files

  • scripts/publish_sticker.py — main deterministic implementation
  • references/api-notes.md — concise implementation notes and known behavior
  • README.md — bilingual human-facing overview for repository/ClawHub readers
  • LICENSE — MIT license

When to read references

Read references/api-notes.md if you need:

  • the exact API mapping (add_material + draft/add)
  • encoding gotchas
  • the distinction between news and newspic

Comments

Loading comments...