WeChat Share

wechat-share

Use this skill when the user wants to move selected files from one OpenClaw workspace to another.

This skill solves two scenarios:

1. Export selected workspace files into a one-time db9 share. 2. Import a received share back into the current workspace.

Supported commands:

Recipients should install wechat-share first, then use preview and import.

Language

What This Skill Shares

Share only files from the current workspace. Typical examples:

Do not share:

Core Model

For each share, return one standard recipient path:

The share instruction must carry enough information for the recipient to install the skill, preview the share, and import the files.

Safety Rules

Required Tools

This skill requires:

If either tool is missing, stop and explain what is missing.

Action Routing

When invoked:

1. If args start with export, follow the export workflow. 2. If args start with import, follow the import workflow. 3. Otherwise ask which action the user wants.

Export Workflow

Step 1: Ask what to share

Ask:

Keep the question short. Offer common candidates if helpful.

Step 2: Validate the selection

For each selected path:

If nothing valid remains, stop and explain why.

Step 3: Confirm before upload

Before creating the share, show:

Step 4: Create the share

Prepare a fresh share container and capture the values needed for import.

1. Create an anonymous db9 account:

curl -sS -X POST "https://api.db9.ai/customer/anonymous-register" \
  -H "Content-Type: application/json" \
  -d '{}'

2. Extract token from the JSON response.

3. Create a database with a unique name such as wechat-share-YYYYMMDD-HHMMSS-RAND:

curl -sS -X POST "https://api.db9.ai/customer/databases" \
  -H "Authorization: Bearer " \
  -H "Content-Type: application/json" \
  -d '{"name":"wechat-share-20260323-103015-ab12"}'

Capture at least:

Step 5: Build the manifest

Build this JSON object with Python:

{
  "version": 1,
  "createdAt": "2026-03-23T10:00:00Z",
  "burnAfterRead": true,
  "items": [
    {
      "remotePath": "/files/SOUL.md",
      "destPath": "SOUL.md",
      "sha256": "..."
    }
  ]
}

Rules:

Recommended manifest shape:

Step 6: Upload files and manifest

Upload every selected file to:

Use the SQL API:

curl -sS -X POST "https://api.db9.ai/customer/databases//sql" \
  -H "Authorization: Bearer " \
  -H "Content-Type: application/json" \
  -d '{"query":"SELECT fs9_write('\"'\"'/files/SOUL.md'\"'\"', '\"'\"'...content...'\"'\"')"}'

Also upload the manifest to /manifest.json with fs9_write().

Do not hand-build large SQL strings in shell. Use Python to:

Step 7: Return the share instructions

Return a localized message that the sender can forward directly.

Always include:

1. A short title 2. A brief preview summary 3. A skill installation step 4. A short /wechat-share preview ... command 5. A short /wechat-share import ... command 6. A short security note

The export result must be structured in two parts:

1. A short sender-facing summary. 2. One complete recipient-facing block that can be copied and forwarded as-is.

Rules for the recipient-facing block:

Label the block clearly, for example:

Use this installation step:

openclaw skills install wechat-share

If openclaw is unavailable but clawhub is available, the fallback is:

clawhub install wechat-share

Use this preview command:

/wechat-share preview --db-id "" --api-token ""

Use this quick command:

/wechat-share import --db-id "" --api-token ""

If burn-after-read is enabled, use:

/wechat-share import --db-id "" --api-token "" --burn

The preview summary should include:

Inside the recipient-facing block, present the flow in this order:

1. Install wechat-share 2. Preview the share 3. Import the share

Aim for a message that feels easy to forward in one shot in a chat app. Keep the wording minimal. Do not add extra narrative before or after the core steps. Use preview-first wording such as:

Step 8: Burn behavior

If burn is requested, prefer deleting the whole share database:

curl -sS -X DELETE "https://api.db9.ai/customer/databases/" \
  -H "Authorization: Bearer "

If whole-database deletion fails, fall back to SQL:

SELECT fs9_remove('/manifest.json');
SELECT fs9_remove('/files', true);

Import Workflow

Step 1: Parse the share

Accept:

If either --db-id or --api-token is missing, ask the user to paste the full share command.

Step 2: Read the manifest

Use the SQL endpoint with curl:

curl -sS -X POST "https://api.db9.ai/customer/databases//sql" \
  -H "Authorization: Bearer " \
  -H "Content-Type: application/json" \
  -d '{"query":"SELECT fs9_read('\"'\"'/manifest.json'\"'\"') AS content"}'

Use Python to parse the JSON response and extract the first cell from rows.

Step 3: Validate the manifest

For every manifest item:

If validation fails, stop before writing anything.

Step 4: Show a preview before import

Before copying any file, show:

If the current action is preview, stop after this step.

Step 5: Download to a temp directory

Never write directly into the workspace before validation and checksum checks finish.

For each manifest item:

Step 6: Copy into the workspace

After all files pass validation:

Then report which files were imported.

Step 7: Burn if requested

Burn should happen when either of these is true:

Preferred burn:

Fallback burn:

Tell the user which burn path succeeded.

Step 8: Return the result

Remind the user:

Implementation Notes

UX Rules For Returned Messages

Dependency Fallback

If curl or python3 is missing:

Do not recommend db9 CLI or psql as required dependencies for this skill.

Examples

For localized share messages and install-first recipient instructions, read examples.md.

For localized share messages and install-first recipient instructions, read examples.md.