Doctor and Inspect Readiness

md2wechat 2.4.0 gives agents a clearer way to decide what is ready before they convert, upload, or create a WeChat draft.

The short rule is:

use doctor --json to check local setup, then use inspect --json to check one article.

What each command answers

doctor is local-only. It checks whether the current machine looks configured enough to attempt conversion, image generation, or WeChat draft work.

inspect checks a specific Markdown file and reports whether the requested targets are ready, blocked, degraded, or not requested.

Recommended order

For a new runtime or upgraded install, start here:

md2wechat version --json
md2wechat doctor --json
md2wechat capabilities --json

For an article workflow, continue with:

md2wechat inspect article.md --json
md2wechat preview article.md
md2wechat convert article.md

For draft creation, inspect the draft target before creating the draft:

md2wechat inspect article.md --draft --cover cover.jpg --json
md2wechat convert article.md --draft --cover cover.jpg

The readiness contract

New agents should prefer these fields from inspect --json:

{
  "data": {
    "readiness": {
      "schema_version": "1",
      "targets": {
        "preview": "ready",
        "convert": "ready",
        "upload": "not_requested",
        "draft": "blocked"
      },
      "blockers": [
        {
          "code": "MISSING_COVER",
          "target": "draft"
        }
      ]
    }
  }
}

The older fields are still available:

• convert_ready
• upload_ready
• draft_ready
• preview_fidelity

But targets and blockers are the better surface for agents because they explain what is blocked and where.

Common blockers

Safe agent behavior

When building an agent workflow, do this:

1. Run doctor --json before API-dependent work.
2. Run the smallest discovery commands needed for the task.
3. Run inspect --json for the article.
4. If a target is blocked, stop that target and report the blocker.
5. Preview before conversion or draft creation when the user has not seen the output.

This keeps automation from guessing and makes failures easier to explain.

Continue with

• Discovery-First Workflow
• Mode Selection
• Article Draft