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.
npx skills add theam/claude-dev-kit --skill plan-definitiongit clone --depth 1 https://github.com/theam/claude-dev-kitWrote 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.
[](https://agentmods.dev/skills/theam/claude-dev-kit/plan-definition)<a href="https://agentmods.dev/skills/theam/claude-dev-kit/plan-definition"><img src="https://agentmods.dev/badge/skills/theam/claude-dev-kit/plan-definition/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.
<a href="https://agentmods.dev/skills/theam/claude-dev-kit/plan-definition"><img src="https://agentmods.dev/badge/skills/theam/claude-dev-kit/plan-definition.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00096 | $0.02889 |
| Opus 5.5 | $0.00038 | $0.01156 |
| Sonnet 5.5 | $0.00019 | $0.00578 |
| Haiku 4.5 | $0.00010 | $0.00289 |
Grade A, and why
plan-definition 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 6d 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.
How it starts
The opening of the file, as written. The whole thing — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Plan Definition
Turn a spark into a clear product definition through a guided discovery phase — the most upstream step of the workflow, for the product-owner persona. It runs before plan-backlog and closes the loop define → backlog → ticket → PR.
plan-backlog starts from a defined problem and turns it into a backlog; its framing step is deliberately light ("how do we slice this"). plan-definition does the deep part: defining the problem itself — who it's for, the outcome, why now, the constraints, the success metrics, the chosen direction. Its output is a definition doc, not a backlog and not tickets. You facilitate the PO's thinking — ask, offer options, and let them decide; never decide the product for them.
Trigger
A request to define, frame, scope, or think through a product idea before there are stories — e.g. "help me define this feature", "I have a rough idea, let's shape it", "what's the problem we're solving here". If the request is already a defined problem asking for tickets, that's plan-backlog, not this.
Boundary with plan-backlog (important)
plan-definition= define the problem & direction. No backlog, no tickets. Output: a product definition.plan-backlog= turn a defined problem into a backlog. When a definition from here is handed in, it consumes it and runs a lighter framing (confirm, don't re-derive) instead of framing from scratch.
1. Intake — read the spark in whatever form it arrives
The input is deliberately vaguer than plan-backlog's — a one-liner is a valid start.
- Pasted text / chat description → use directly.
- PDF → read it (page range as needed).
- Word (
.docx) → convert first (textutil -convert txt file.docx -output -on macOS, orpandoc file.docx -t markdown), then read. If neither tool is available, ask the user to paste the text or export a PDF. - Artifact / Confluence page / URL → fetch it.
- Figma link (
figma.com/(design|file)/…) → runfigma-fetchfor design context. For a FigJam board,figma-fetchdoesn't read/board/URLs yet — ask the PO to paste the board's content as text. v1 treats any whiteboard as read-only context; nothing is written back to it.
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.
- 6d ago First seen · 112 lines · 96 tokens per session scan A 2c8e99e60be4
plan-definition is a skill published in the GitHub repository theam/claude-dev-kit (14 stars, last pushed yesterday), licensed Apache-2.0. It adds 96 tokens to every session and 2,889 once invoked, about $0.0004 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 2026-09-24.
Other skills, from other repositories
protocol-tracker
Internal skill for commands. The tracker verb-resolution protocol - how a lets-tracker block resolves through the active adapter, the neutral read shape, body crossing, degradation, preflight and trust. Loaded by a command or skill right before its first tracker verb. Do not trigger on user conversation.
settle
Retire a released work item by folding what is still true out of its spec.md into a docs/ policy document, then removing the directory. Use when: preparing a release and the work items have accumulated; a session asks what seal/specs/ is still holding, or runs settle. NOT for: gathering changelog or ledger fragments…
x-dev
A development workflow for completing tasks from a task folder. It follows the task's README and checklist, implements work in dependency order, and records verification evidence.
x-req
A requirements planner that turns a confirmed request into a ready-to-build task package. It assigns a risk level from Q0 to Q3, where higher levels cover changes such as permissions, public interfaces, data migrations, or concurrency.
triage-task
Reads a ticket from Jira or Azure DevOps and writes .workflow/ /triage.json. Use when asked what to work on, to pick up a ticket by key, or to understand its description, comments and blocking links.
deliver-task
Drive one READY task end-to-end — implement in a worktree, independent adversarial review, bounded 3-iteration patch loop, then hand to a human for merge. Pass the issue number, e.g. "/deliver-task 42". A single-command composition of next-task + review-task + patch-task.