shipkit-stage

A project-stage guide that labels work as a proof of concept, alpha, MVP, or scale and defines the quality checks for moving forward. An MVP is the smallest usable version of a product.

In plain words
What is it for?
Setting a project's stage, defining constraints and graduation criteria, and evaluating whether it is ready for human approval.
Why use it?
It turns a vague sense of progress into documented scope limits, evidence, business measures, and approval gates, and pauses when the stage cannot be grounded in evidence.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/stefan-stepzero/shipkit/shipkit-stage
Any agent
npx skills add stefan-stepzero/shipkit --skill shipkit-stage
Clone the repo
git clone --depth 1 https://github.com/stefan-stepzero/shipkit

Made for: Claude Code, Codex.

Per session 30 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,303 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00030 $0.05303
Opus 5 $0.00015 $0.02652
Sonnet 5 $0.00006 $0.01061
Haiku 4.5 $0.00003 $0.00530

Measured 2d ago against content hash db5406ff82bc, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

shipkit-stage 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 2d 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.

install/skills/shipkit-stage/SKILL.md · 465 lines

How it starts

The opening of the file, as written. The whole thing — 465 lines — stays where its author put it; the contents beside it link to each section on GitHub.

shipkit-stage — Project Stage & Graduation Criteria

Purpose: Define the project stage (POC/Alpha/MVP/Scale), derive scope constraints and quality bars from that stage, set business-metric criteria (S-*), and define stage gates. Evaluate mode assesses gate readiness for human-approved graduation.

What it does: Reads the project vision and codebase signals, grounds the stage from cited signals, derives constraints and business metrics, defines gates with S-* criteria, and writes goals/strategic.json. Runs in fork context — when stage is grounded by cited signals, proceeds autonomously; when stage is genuinely ungrounded, emits NEEDS_ELICITATION:shipkit-stage and pauses rather than guessing silently. Evaluate mode is user-invoked and human-gated — it cross-references ALL goal files to build a graduation evidence table for human approval.

Protocol: This skill follows the canonical elicitation protocol defined in install/shared/references/elicitation-protocol.md (the mechanics — marker, state files, resume). The steps below are this skill's specific application of that protocol.

Calibration: Apply install/shared/references/ground-or-ask-calibration.md (the intelligence — propose vs ask). Ground first: propose every stage field you can tie to a cited signal (the opening prompt, why.json stage/currentState/approach/constraints keywords, codebase maturity signals — file counts, CI/CD config, specs, tests, deployment infrastructure), tagged with its source; flag low-leverage guesses as guessed. For shipkit-stage the high-leverage decisions are: (1) the stage itself (POC / Alpha / MVP / Scale) when no cited signal resolves it, and (2) graduation scope constraints when stage is genuinely ambiguous. These set hard scope boundaries — a wrong stage silently misaligns every constraint, quality bar, and success criterion derived from it. When stage is grounded by a clear cited signal, proceed autonomously (no marker). Only emit NEEDS_ELICITATION:shipkit-stage when stage is genuinely ungrounded after the full inference pass (no explicit stage field, no unambiguous keywords, no codebase maturity signal points clearly at one stage).

Output: One JSON file:

  • goals/strategic.json — Stage, constraints, stageImplications, business-metric criteria, gates

Product goals (user outcomes: P-) are handled by /shipkit-product-goals. Engineering goals (technical performance: E-) are handled by /shipkit-engineering-goals. Both skills append their criteria IDs to the gates defined here.


Read the full file on GitHub · 465 lines

Files

What ships with it

2 files 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. 2d ago First seen · 465 lines · 30 tokens per session scan A db5406ff82bc

Subscribe to this mod's changes

shipkit-stage is a skill published in the GitHub repository stefan-stepzero/shipkit (1 stars, last pushed 1mo ago), licensed MIT. It adds 30 tokens to every session and 5,303 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.