shipyard-disciplined-builder

shipyard-disciplined-builder is an agent for Claude Code from Acendas/shipyard. It costs 105 tokens per session (4,223 once invoked), scanned A, original, MIT.

A disciplined implementer for one Shipyard task in an isolated worktree, which is a separate working copy of the repository. It follows TDD, or test-driven development: write a failing test, add the smallest implementation, run checks, scan for unfinished code, commit, and report the result.

In plain words
What is it for?
Use it to implement a single Shipyard feature task, verify its acceptance probe, commit the changes, and return a structured status.
Why use it?
It gives each task a repeatable implementation process and helps prevent untested or placeholder code from being committed.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions subagents.

Part of the shipyard plugin — 32 skills, 7 agents, 4 hooks shipped together

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 agents/acendas/shipyard/shipyard-disciplined-builder
Clone the repo
git clone --depth 1 https://github.com/Acendas/shipyard

Made for: Claude Code.

Or install shipyard, the plugin that ships this one along with the rest of its 32 skills, 7 agents, 4 hooks.

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 shipyard-disciplined-builder

README.md
[![agentmods](https://agentmods.dev/badge/agents/acendas/shipyard/shipyard-disciplined-builder.svg)](https://agentmods.dev/agents/acendas/shipyard/shipyard-disciplined-builder)
Your own site
<a href="https://agentmods.dev/agents/acendas/shipyard/shipyard-disciplined-builder"><img src="https://agentmods.dev/badge/agents/acendas/shipyard/shipyard-disciplined-builder.svg" alt="Measured on agentmods" height="20"></a>
Per session 105 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,223 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 5.1 $0.00105 $0.04223
Opus 5 $0.00053 $0.02112
Sonnet 5 $0.00021 $0.00845
Haiku 4.5 $0.00011 $0.00422

Measured 5d ago against content hash 0502edc250cf, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

shipyard-disciplined-builder 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 5d 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.

plugins/shipyard/agents/shipyard-disciplined-builder.md · 280 lines

How it starts

The opening of the file, as written. The whole thing — 280 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Shipyard Disciplined Builder

You are executing one Shipyard sprint task in an isolated subagent context, using the brief the orchestrator gave you in this prompt (task ID, working branch, worktree path, base ref, data dir, task file path, feature file path, sprint ID, wave number). If the brief is missing any required parameter — most critically task_id, task_file_path, acceptance_probe, or data_dir — stop immediately and return:

STATUS: BLOCKED
ESCALATION_CODE: (omit — no code fits a missing-brief failure)
REASON: <name the missing parameter(s)>

Otherwise, proceed.

Environment & rules (read before your first action)

  1. Worktree branch self-check — your VERY FIRST action. This check depends on the Isolation: line in your brief.

    • Isolation: on (or the line is absent — the isolated default): run git branch --show-current. It MUST match shipyard/wt-*. If it does not, you are NOT in your isolated worktree — STOP immediately and return STATUS: BLOCKED with ESCALATION_CODE: isolation_failure. Do NOT "fix" this by checking out the working branch yourself; that bypasses isolation and races the other builders in this wave.
    • Isolation: off (in-place mode — the orchestrator dispatched you sequentially on the working branch, no worktree): SKIP the shipyard/wt-* assertion. Instead confirm you are on the Working branch from your brief; if not, STOP with ESCALATION_CODE: isolation_failure (a wrong branch here is still a real dispatch error). In this mode you are the only builder running, so committing on the working branch is correct and expected — but you MUST still make exactly one atomic task commit, never leave the tree dirty.
  2. Kind refusal. This loop is for feature tasks only. Read the task file frontmatter first: if kind: operational or kind: research, STOP and return STATUS: BLOCKED with ESCALATION_CODE: misrouted_kind. Those kinds have different deliverables and different dispatchers — do not attempt them here.

Read the full file on GitHub · 280 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. 5d ago First seen · 280 lines · 105 tokens per session scan A 0502edc250cf

Subscribe to this mod's changes

shipyard-disciplined-builder is an agent published in the GitHub repository Acendas/shipyard (2 stars, last pushed 24d ago), licensed MIT. It adds 105 tokens to every session and 4,223 once invoked, about $0.0005 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 agents, from other repositories

ap-implementer

L3 executor - G4 IMPLEMENT. Builds one feature from its approved executable roadmap item or conditional frozen plan using strict TDD and real test runs; coverage >=95% on changed lines. Reports PLAN-CONFLICT rather than improvising.

Spielewoy/autoprompt-skill · 52 tokens

test-writer

Use for generating comprehensive tests following TDD/BDD principles.

FlorianBruniaux/claude-code-ultimate-guide · 15 tokens

doncheli

Don Cheli SDD Framework — Specification-Driven Development with TDD iron law, 15 reasoning models, and 93+ commands. Autonomous Mode, OWASP audit, PRD Generator, Crash Recovery. Activate when user mentions: start, specify, implement, review, estimate, debate, plan, audit, migrate, test, spec, TDD, or any Don Cheli…

doncheli/don-cheli-sdd · 79 tokens

builder

Implements features, fixes bugs, and creates projects from specs. Follows TDD, uses conventional commits, and prefers boring solutions. Delegated from /ccc-build after…

KevinZai/commander · 37 tokens

tdd-coach

TDD red-green-refactor 사이클 코치. 테스트 먼저 작성을 강제하고 커버리지를 확인. "TDD", "테스트 먼저", "test first" 요청에 실행.

Dannykkh/skill-olympus · 46 tokens

backend-phase-5

You are the Service Layer Agent. You build the business logic layer using test-driven development. You write tests FIRST with mocked repositories, then implement services to make them pass. Services orchestrate domain logic and transform data between domain and API types.

TouheedCode/claude-dev-workflow · 0 tokens