what-next

what-next is a skill for Claude Code, Codex from quanghoangf/vibedoc. It costs 166 tokens per session (1,187 once invoked), scanned A, original, MIT.

A read-only project-status helper that recommends one most useful next action and gives the exact command to run.

In plain words
What is it for?
Use it when you ask what to do next in a VibeDoc project, especially when work may be blocked, abandoned, or ready to resume.
Why use it?
It removes the need to inspect tasks, roadmap entries, activity, handoff notes, and Git state yourself or weigh a long list of options.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one.

Good fit Use it when you ask what to do next in a VibeDoc project, especially when work may be blocked, abandoned, or ready to resume.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/quanghoangf/vibedoc/what-next
Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent
npx skills add quanghoangf/vibedoc --skill what-next
Clone the repo
git clone --depth 1 https://github.com/quanghoangf/vibedoc

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for what-next

README.md
[![agentmods](https://agentmods.dev/badge/skills/quanghoangf/vibedoc/what-next/github.svg)](https://agentmods.dev/skills/quanghoangf/vibedoc/what-next)
Your own site
<a href="https://agentmods.dev/skills/quanghoangf/vibedoc/what-next"><img src="https://agentmods.dev/badge/skills/quanghoangf/vibedoc/what-next/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for what-next

Your own site · 80×15
<a href="https://agentmods.dev/skills/quanghoangf/vibedoc/what-next"><img src="https://agentmods.dev/badge/skills/quanghoangf/vibedoc/what-next.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 166 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,187 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00166 $0.01187
Opus 5.5 $0.00066 $0.00475
Sonnet 5.5 $0.00033 $0.00237
Haiku 4.5 $0.00017 $0.00119

Measured yesterday against content hash 1873325b3005, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-30, from the pricing page.

Security

Grade A, and why

what-next scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/what-next/SKILL.md · 53 lines

How it starts

The opening of the file, as written. The whole thing — 53 lines — stays where its author put it; the contents beside it link to each section on GitHub.

What next

Answer "what should I do now?" with one recommendation and a command the user can run immediately. A list of ten options is not an answer. The user asked because they didn't want to weigh options themselves. Do the weighing, show the reasoning in a line or two, and keep the alternatives short.

This skill only reads. Don't change task or roadmap files, and don't start the recommended work until the user says so.

1. Gather the state

Read these, in any order. Prefer the VibeDoc MCP tools when they're connected (vibedoc_get_status, vibedoc_get_roadmap, vibedoc_list_tasks). Otherwise read the files directly. Either way it's cheap, so read everything below before deciding.

  • Tasks: plans/tasks/T*.md. For each: id, title, status, Depends on, and Due from the **Key:** Value block under the H1.
  • Roadmap: plans/roadmap/R*.md. Items with no **Parent:** are horizons, and items with one are epics. For each epic: status, **Tasks:** list, and Due.
  • Activity: .vibedoc-activity.json. You need the last task_updated time for each in-progress task, to tell active work from abandoned work.
  • Handoff: the "Handoff for next session" / "Up next" part of memory/MEMORY.md, if it exists. It is the last session's intent. Weigh it, but check it against the files, because it goes stale.
  • Git: current branch, uncommitted changes, and commits not pushed yet (git status -sb, git log @{u}.. when there is an upstream).

2. Pick the first rule that applies

The order runs from "something is stuck or at risk" to "start something new". Unblocking beats starting, because a stuck task holds up everything that depends on it, and starting new work on top of a mess makes the mess bigger.

# Situation Recommend
1 Uncommitted changes, or a branch with unpushed commits, that look like finished work Commit or push / open a PR (/commit, /create-pr). Say what the changes are.
2 A task is blocked, or an epic's queue would report "Needs a human" (blocked task, missing task file, broken dependency) Unblock it: read its ## Blocked because note if there is one and say what decision is needed.
3 A task is in-progress with no activity for more than ~24h It's probably abandoned: resume it, or move it back to todo so the queue can hand it out again.
4 An epic has ready tasks: todo, with every dependency done or cancelled /work-epic <epic>. Prefer an epic that is already in-progress, then the earliest horizon, then by Order. Name the task it would start with.
5 Roadmap drift: all of an epic's tasks are done but the epic isn't, an epic is planned while its tasks are moving, or something is overdue Fix the status or due date (vibedoc_update_roadmap_item, or edit the **Status:** line).
6 The earliest horizon has a planned epic with no tasks /epic-breakdown <epic>. Pick the first one by horizon order, then by Order.
7 No roadmap at all (plans/roadmap/ empty or missing) /roadmap-planner.
8 Nothing above applies Say the project is in a clean state, and suggest /roadmap-planner to plan the next horizon.

Read the full file on GitHub · 53 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 53 lines · 166 tokens per session scan A 1873325b3005

Subscribe to this mod's changes

what-next is a skill published in the GitHub repository quanghoangf/vibedoc (5 stars, last pushed today), licensed MIT. It adds 166 tokens to every session and 1,187 once invoked, about $0.0007 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-29.