smithery.ai

vd-next

Advance a VibeDev workflow from chat. Use when the user types `$vd-next` / says “vd next” and wants the agent to fetch the next compiled step prompt from the VibeDev MCP HTTP server (`vibedev-mcp serve`) and handle breakpoint/new-thread/diagnose states.

First seen Mar 25, 2026

Installation

$ npx skills add https://smithery.ai

Similar popular skills

Related neighbors and high-traction skills in the same topics — useful to compare before installing.

Also in this package

Other skills from smithery.ai · top by installs.

npx skills add https://smithery.ai

Browse all from smithery.ai

More details

Agent compatibility

Declared targets from SKILL.md / docs. Unmarked agents are not listed — the skill may still install via the CLI.

Claude Code Not declared
Cursor Not declared
Codex Not declared
GitHub Copilot Not declared
Windsurf Not declared
Gemini CLI Not declared
Cline Not declared
OpenCode Not declared

Package contents

Files included with this skill beyond the listing page.

  • skill md SKILL.md 1,540 B
  • docs SUMMARY.md 272 B

History

  1. First seen on skills.sh
  2. First recorded snapshot · 1 installs

SKILL.md

vd-next (VibeDev Next Step Driver)

Goal

Make “type $vd-next repeatedly” a reliable workflow:

  • fetch the next step prompt from the VibeDev server
  • respect NEWTHREAD / AWAITHUMAN / DIAGNOSE
  • print the exact prompt to run next (with the server’s completion marker)

How to run (scripted)

Run the bundled script:

  • python skills/vd-next/scripts/vd_next.py

Common options:

  • --job-id <JOB_...> (if omitted, picks the first EXECUTING job)
  • --server http://127.0.0.1:8765
  • --json (print raw JSON response)

Expected loop

  1. User calls $vd-next
  2. Fetch next action from /api/jobs/{job_id}/next-prompt-auto
  3. Output:

- If action=NEXTSTEP or RETRY: print the prompt to execute - If action=NEWTHREAD: instruct user to start a new thread and print the seed prompt - If action=AWAIT_HUMAN or DIAGNOSE: stop and tell user what to do

  1. After the step is done, the agent (or user) submits evidence via VibeDev tools / UI
  2. Repeat

Notes:

  • Breakpoints are implemented as a normal step that produces a Memory Pack and sets a NEW_THREAD flag for the following step.
  • Memory Packs should be saved as context blocks tagged carryforward (and optionally carrytostep:<stepid>).