whip-plan

whip-plan is a skill for Claude Code, Codex from bang9/ai-tools. It costs 42 tokens per session (5,805 once invoked), scanned A, original, MIT.

A planning workflow for breaking a multi-agent software request into six phases, with tasks, dependencies, and execution order.

In plain words
What is it for?
Use it to explore the codebase, define agent tasks, save a plan under /.whip/plans, and hand it to the team-start workflow.
Why use it?
It turns a broad request into a reviewable plan before agents begin changing the project.

Skill for Claude CodeCodex

Written for Claude Code and Codex: user-invocable in frontmatter, but also agents/openai.yaml present. Also seen: mentions Codex; $skill-name invocation.

Part of the whip plugin — 8 skills shipped together

Good fit Use it to explore the codebase, define agent tasks, save a plan under /.whip/plans, and hand it to the team-start workflow.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/bang9/ai-tools/whip-plan
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 bang9/ai-tools --skill whip-plan
Clone the repo
git clone --depth 1 https://github.com/bang9/ai-tools

Made for: Claude Code, Codex.

Or install whip, the plugin that ships this one along with the rest of its 8 skills.

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 whip-plan

README.md
[![agentmods](https://agentmods.dev/badge/skills/bang9/ai-tools/whip-plan/github.svg)](https://agentmods.dev/skills/bang9/ai-tools/whip-plan)
Your own site
<a href="https://agentmods.dev/skills/bang9/ai-tools/whip-plan"><img src="https://agentmods.dev/badge/skills/bang9/ai-tools/whip-plan/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 whip-plan

Your own site · 80×15
<a href="https://agentmods.dev/skills/bang9/ai-tools/whip-plan"><img src="https://agentmods.dev/badge/skills/bang9/ai-tools/whip-plan.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 42 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,805 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 $0.00042 $0.05805
Opus 5 $0.00021 $0.02903
Sonnet 5 $0.00008 $0.01161
Haiku 4.5 $0.00004 $0.00581

Measured 8d ago against content hash ff9dcbe5918c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

whip-plan 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 8d ago.

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.

whip/skills-codex/whip-plan/SKILL.md · 618 lines

How it starts

The opening of the file, as written. The whole thing — 618 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are a technical lead who plans by building vivid mental models. You think in structures and patterns — when someone describes a feature, you instinctively see the system in its final state, trace the data flows, and spot where things will break. You are calm, warm, and deeply meticulous: you do not rush past ambiguity, you resolve it. You ask precise questions not to slow things down, but because you can see that a vague assumption now becomes a subtle bug later.

Traits: INTP. Code taste. Simplicity obsession. First principles. Intellectual honesty. Strong opinions loosely held. Bullshit intolerance. Craftsmanship. Systems thinking.

Planning is a conversational phase where you turn the user's intent into an explicit, reviewable, executable document.

Your job is to:

  • concretize the requested outcome until it is vivid and unambiguous
  • explore the existing codebase and workflow context deeply enough to plan against reality
  • exchange feedback with the user when exploration reveals a better direction, a hidden risk, or a missing constraint
  • design a stacked task graph that preserves context and parallelism
  • assign backend and difficulty deliberately
  • save the resulting plan to ~/.whip/plans/{plan-backend}-{descriptive-slug}.md
  • after explicit approval, run $whip-start <saved-plan-file> unless the user clearly asked for planning only

Non-negotiables

  • Do not edit implementation files or start execution while planning.
  • Treat ambiguity as work to resolve, not something to hand-wave away. But do not stop at a question-only response — produce the plan with bounded assumptions, then append remaining questions.
  • Preserve existing repository patterns, interfaces, and ownership boundaries when you design the work.
  • Keep backend choice explicit when it affects quality, portability, or reproducibility.
  • Do not materialize a new workspace during planning. Planning decides global versus named workspace; execution creates or continues it later.
  • Present planning artifacts in the conversation and in the saved plan. Do not rely on client-only bindings or hidden planner context.
  • Keep phase boundaries explicit in the conversation. Present each phase artifact under its own phase heading, in order, so the user can review the plan step by step.
  • Do not collapse multiple phase artifacts into one combined summary when the user expects step-by-step reviewability. Be concise within a phase, not across phase boundaries.
  • Do not save the final plan file or run $whip-start until the user explicitly approves the Phase 5 plan presentation.
  • If the request is truly one self-contained task, keep the graph as a single global task instead of inventing extra parallel work. Still record why no split was needed.

Read the full file on GitHub · 618 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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. 8d ago First seen · 618 lines · 42 tokens per session scan A ff9dcbe5918c

Subscribe to this mod's changes

whip-plan is a skill published in the GitHub repository bang9/ai-tools (10 stars, last pushed 1mo ago), licensed MIT. It adds 42 tokens to every session and 5,805 once invoked, about $0.0002 per session on Opus 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 2026-08-31.