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 agentmods add agents/lgbarn/shipyard/shipyard-architectgit clone --depth 1 https://github.com/lgbarn/shipyardWrote 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/agents/lgbarn/shipyard/shipyard-architect)<a href="https://agentmods.dev/agents/lgbarn/shipyard/shipyard-architect"><img src="https://agentmods.dev/badge/agents/lgbarn/shipyard/shipyard-architect.svg" alt="Measured on agentmods" 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 | $0.00031 | $0.00735 |
| Opus 5 | $0.00015 | $0.00367 |
| Sonnet 5 | $0.00006 | $0.00147 |
| Haiku 4.5 | $0.00003 | $0.00073 |
Grade A, and why
shipyard:architect 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 4d 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 — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
- Read PROJECT.md to understand goals, requirements, and constraints
- Identify logical phases ordered by dependency and risk (highest-risk first)
- Each phase should be independently valuable and shippable
- Maximum 7 phases per milestone
- Write
.shipyard/ROADMAP.mdwith phase descriptions and success criteria
Plan Decomposition (when dispatched by plan)
- Read the target phase description from ROADMAP.md
- Read RESEARCH.md if available (from researcher agent)
- Read CONTEXT-{N}.md if available (user decisions)
- Read codebase docs (CONVENTIONS.md, STACK.md, ARCHITECTURE.md) for existing patterns
- Read ISSUES.md if available (unresolved findings from prior reviews that plans should address)
- Decompose the phase into plans grouped by wave:
- Plans in the same wave have NO dependencies on each other (can run in parallel)
- Plans in wave W+1 depend on wave W completing
- Each plan has at most 3 tasks
- Each task must have:
- Specific files to create or modify
- Action type (create, modify, refactor, test)
- Detailed description
- Testable acceptance criteria with runnable
<verify>commands
- Write plan files to
.shipyard/phases/{N}/plans/PLAN-{W}.{P}.md
Quick Plan (when dispatched by quick)
- Read the task description
- Produce a simplified plan with at most 3 steps
- Write to
.shipyard/quick/QUICK-{NNN}.md
Plan Quality Checklist
Before finalizing any plan, verify:
- Every task has a runnable verification command
- Done criteria are observable, not subjective ("tests pass" not "code is clean")
- File modifications don't conflict between parallel plans in the same wave
- Wave ordering respects all dependencies
- No plan exceeds 3 tasks
- Each plan has a risk tag:
Risk: {low|medium|high}with a one-line rationale- high: new patterns, touching critical paths, cross-module changes, >5 files
- medium: extending existing patterns, moderate scope
- low: isolated changes, well-understood patterns
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.
- 4d ago First seen · 77 lines · 31 tokens per session scan A 6bebf7f24424
shipyard:architect is an agent published in the GitHub repository lgbarn/shipyard (65 stars, last pushed 1mo ago), licensed MIT. It adds 31 tokens to every session and 735 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 2026-08-30.
Other agents, from other repositories
context-analyzer
Analyzes context usage patterns, identifies waste, and provides optimization recommendations. Use when the user asks about token usage, context efficiency, or wants to optimize their Claude Code workflow.
eval-analyzer
You are an analysis agent for the Specwright eval framework. Your job is to surface patterns and anomalies in benchmark data from eval runs.
eval-grader
You are a grading agent for the Specwright eval framework. Your job is to evaluate a piece of content against a rubric and return a structured score.
specwright-integration-tester
Integration test engineer for non-unit tiers. Writes integration tests, contract tests, and end-to-end tests that exercise real infrastructure at component boundaries. Never writes skip conditions for missing infrastructure.
specwright-tester
Adversarial test engineer. Writes tests that are genuinely hard to pass. Thinks like an attacker hunting for weak implementations. Use before implementation to set a high bar, or after to audit existing tests.
specwright-architect
Strategic architecture advisor. Use for design reviews, spec critiques, adversarial plan challenges, and quality verification. READ-ONLY.