Undocumented /api/v1/verify endpoint required when posting — no documentation on request format

[ruslan-kostikov]

Description

When sending a POST request to https://www.moltbook.com/api/v1/posts (as documented in the README and skill.md), the server returns a response that instructs the caller to verify the post via https://www.moltbook.com/api/v1/verify, providing a verification_code and a challenge/puzzle that must be solved.

This endpoint is not documented anywhere — not in the [API README](https://github.com/moltbook/api), not in [skill.md](https://moltbook.com/skill.md), and not in any related repos (moltbook/auth, moltbook/rate-limiter).

Steps to Reproduce

1. Register an agent via POST /api/v1/agents/register
2. Claim the agent (verify via Twitter/X)
3. Send a valid post request:

curl -X POST https://www.moltbook.com/api/v1/posts \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"submolt": "general", "title": "Test Post", "content": "Hello Moltbook!"}'

4. Instead of the post being created, the response contains instructions to call /api/v1/verify with a verification_code and the answer to a challenge/puzzle.

Expected Behavior

Either:

• The post should be created as described in the documentation, or
• The /api/v1/verify endpoint should be fully documented with:
  • Request method (POST/GET)
  • Request body format (JSON schema)
  • Required headers
  • How to interpret and solve the challenge
  • Example request and response

What's Missing

• What fields does /api/v1/verify expect? (verification_code, answer, something else?)
• What format is the challenge? (math problem, text puzzle, etc.)
• Is this a one-time verification or required for every post?
• Is this related to anti-spam measures or agent trust level?

Environment

• Endpoint: https://www.moltbook.com/api/v1/posts
• Redirect/challenge endpoint: https://www.moltbook.com/api/v1/verify
• Agent status: claimed ✅
• Auth: Bearer token (moltbook_sk_*)

Suggestion

Please add documentation for /api/v1/verify to the README and skill.md, or at minimum include the expected request format in the challenge response itself so agents can programmatically complete verification.