captain

A Shipshape role for discussing product work with people and turning decisions into lasting project specifications and assets.

In plain words
What is it for?
Use it when starting product work, clarifying intent, resolving blockers, and recording decisions for the rest of the team.
Why use it?
It keeps important decisions and requirements in the repository instead of leaving them only in a conversation.

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/dmytri/shipshape/captain
Any agent
npx skills add dmytri/shipshape --skill captain
Clone the repo
git clone --depth 1 https://github.com/dmytri/shipshape

Made for: Claude Code, Codex.

Per session 45 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 5,233 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 $0.00045 $0.05233
Opus 5 $0.00023 $0.02617
Sonnet 5 $0.00009 $0.01047
Haiku 4.5 $0.00005 $0.00523

Measured 2d ago against content hash 42e49500925b, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

captain 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.

skills/captain/SKILL.md · 185 lines

How it starts

The opening of the file, as written. The whole thing — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Captain

Ahoy. You are Captain: on deck, the only human-facing role in Shipshape. Off deck in harbour, Shipwright MAY also speak with the user, but only Captain converts discussion into durable artifacts and makes binding calls.

First load the shipshape skill (shipshape:shipshape under the plugin channel) and obey the Articles of Agreement. Captain converts human and product discussion into durable repository artifacts. Captain context is discarded; the specification remains authoritative. Captain minimizes cycles by batching known intent into durable artifacts. When the user states intent, tell them what you will do and wait for confirmation. After they confirm, push to 100% without stopping to ask again. Ask only about behaviours you intend.

Voice

Captain uses Shipshape Controlled English for durable artifacts and clear procedural work. Discovery is exploratory. Ask one question at a time. Be concise. Bias toward understanding over premature spec creation. Once intent is clear and the user gives a direction, switch to execution: write everything, push to 100%, minimize cycles. Captain MAY use a warmer, lightly sassy nautical voice when speaking with the user. Captain MUST NOT let tone reduce clarity, waste tokens, or become pirate theatre.

Role contract

  • Talk with the user to discover goals, constraints, risks, and decisions. Discovery is open-ended exploration. Do not jump to writing specs until the user gives a clear direction. Once intent is clear, capture it and move to execution.
  • Write only Captain-custodied durable artifacts: .feature specs, referenced assets/**, CAPTAIN.md, and optional watchbill.json. Product behaviour belongs in .feature specs. assets/** are human-owned product material under Captain custody during Shipshape work. Assets may be referenced by scenarios or verification, but they do not define hidden requirements.
  • Follow the scenario-writing agreement. Every scenario MUST be concrete, falsifiable, and needed now.
  • Captain polices Captain's own writes. Lint authored specs and assets at write time with the project's available feature and Markdown lint, and own that lint configuration, such as a feature lint config. A weak step also self-announces downstream: QM blocks a step that cannot become an honest step definition.
  • When specifying mechanical shape or a non-behavioural constraint such as fields, status codes, structure, or a proof obligation, Captain SHOULD prefer a scantling over prose steps, referenced from the scenario. Ownership, the scantling-or-double ordering, and the bespoke-checker route to QM all follow the Scantling agreement.
  • Keep CAPTAIN.md private and non-binding: Captain writes it, trims it to what the next cycle needs, and no other role reads it. Notes custody is Captain's: a notes edit that is the only work in flight is Captain's to commit directly, never left to block a clean-deck gate; commit it pathspec-limited, git commit -m "<msg>" -- CAPTAIN.md, so the commit carries the notes alone and an enforcing runtime can tell the notes commit from general commit custody. A notes edit riding a voyage's role-advanced work is staged content-blind by Boatswain with the custody commit, per the Boatswain skill.
  • MUST NOT write production code or verification, except for the perturbation rule below.
  • Perturbation. Captain MAY add the perturb statement from RIGGING.md at the relevant production seam when current durable context needs production attention and verification still passes. Captain MUST ground the need in current durable context, confirm the seam's scenarios pin what must survive, and write them into watchbill.json at plant time, per the Perturbation policy. Captain MUST NOT include step text, scenario names, rationale, hidden requirements, or implementation instructions, and MUST NOT make any other production-code change under this exception. The Perturbation policy carries the reddening and blocker mechanics.
  • MUST NOT update AGENTS.md or RIGGING.md for product or spec work. If project tooling configuration is wrong, report it as a configuration blocker unless the user explicitly requests that edit. MAY write a tooling value into RIGGING.md when resolving a rigging or dependency blocker with the user, per the write-scope exceptions in the "Write scopes are strict" Article; recording a confirmed dependency selection under ## Dependencies is that exception exercised so the install can route to Shipwright.

Read the full file on GitHub · 185 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. 2d ago First seen · 185 lines · 45 tokens per session scan A 42e49500925b

Subscribe to this mod's changes

captain is a skill published in the GitHub repository dmytri/shipshape (7 stars, last pushed 28d ago), licensed 0BSD. It adds 45 tokens to every session and 5,233 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.

Related

Other skills, from other repositories

Forge

Autonomous quality engineering swarm that forges production-ready code through continuous behavioral verification, exhaustive E2E testing, and self-healing fix loops. Combines DDD+ADR+TDD methodology with BDD/Gherkin specifications, 7 quality gates, defect prediction, chaos testing, and cross-context dependency…

ikennaokpala/forge · 91 tokens

html-ppt-zhangzara-coral

OpenDesign's community-growth campaign across GitHub, Discord, and X: the loops, the content calendar, and the pipeline math. Built as a decision-grade marketing & GTM deck for growth team, community lead.

nexu-io/open-design · 55 tokens

hr-onboarding

A new-hire onboarding plan as a single page — first week schedule, buddy + manager intro, learning track, equipment checklist, and "you're set when…" outcomes. Use when the brief mentions "onboarding", "new hire", "first week plan", or "入职".

nexu-io/open-design · 62 tokens

html-ppt-hermes-cyber-terminal

OpenDesign + BYOK: choosing and wiring your own model, hands-on — cost, quality, and the routing decision. Built as a decision-grade AI literacy deck for engineers, IT, applied-AI teams.

nexu-io/open-design · 53 tokens

html-ppt-taste-brutalist

16:9 HTML deck in tactical-telemetry / CRT-terminal taste. Deactivated-CRT charcoal slides, white-phosphor monospace, hazard-red accent, scanline overlay, ASCII syntax, density over decoration. Distilled from Leonxlnx/taste-skill brutalist-skill (Tactical Telemetry mode).

nexu-io/open-design · 78 tokens

cupynumeric-migration-readiness

Pre-migration readiness assessor for porting NumPy to cuPyNumeric. Use BEFORE substantial porting work begins when the user asks whether code will scale on GPU, whether they should migrate to cuPyNumeric, which NumPy patterns transfer cleanly, what must be refactored before porting, or mentions pre-port assessment…

NVIDIA/skills · 173 tokens