Borrowing it
Nothing to install: this file belongs to sethdford/shipwright. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/sethdford/shipwright/main/AGENTS.mdgit clone --depth 1 https://github.com/sethdford/shipwrightWrote 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/instructions/sethdford/shipwright/agents-md)<a href="https://agentmods.dev/instructions/sethdford/shipwright/agents-md"><img src="https://agentmods.dev/badge/instructions/sethdford/shipwright/agents-md/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/instructions/sethdford/shipwright/agents-md"><img src="https://agentmods.dev/badge/instructions/sethdford/shipwright/agents-md.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 | $0.01350 | $0.01350 |
| Opus 5.5 | $0.00540 | $0.00540 |
| Sonnet 5.5 | $0.00270 | $0.00270 |
| Haiku 4.5 | $0.00135 | $0.00135 |
Grade C, and why
shipwright AGENTS.md scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
`python -c` with expansions, no `curl | sh`, no `${{`, no `| grep -q`, no BSD `date -j`/`date -r` · F12 dashboard layers, subprocesses only in Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
`python -c` with expansions, no `curl | sh`, no `${{`, no `| grep -q`, no BSD `date -j`/`date -r` · F12 dashboard layers, subprocesses only in How it starts
The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Shipwright — agent guide
Shipwright orchestrates autonomous Claude Code agent teams: delivery pipelines from issue to PR,
a daemon that processes labeled issues, fleet operations, memory, and cost and DORA metrics.
It is being migrated from bash (scripts/) to TypeScript on Bun (src/). Design:
docs/superpowers/specs/2026-10-13-shipwright-ts-migration-design.md.
Definition of done
bun run check:fastafter each change (~10–25 s; a Stop hook runs it too).bun run checkbefore you call work done (< 60 s). CI runs the same command.
A change is done when bun run check passes. Failure messages name the rule and the fix.
Layout (TypeScript)
src/contexts/<ctx>/domain— pure logic. No I/O, no Node/Bun built-ins.src/contexts/<ctx>/application— use cases; depend on ports only.src/contexts/<ctx>/index.ts— the only file other code may import from a context.src/shared/kernel—Result, shared ports and the durability ports (types only).src/adapters— the only code that touches processes, files, network, env, tmux or Claude.src/cli— thin commands;src/main.tswires adapters into use cases.
Target contexts: delivery, orchestration, planning, quality, learning, intelligence, integrations, observability, teams, platform. In code so far: orchestration, learning, integrations, platform.
Fitness rules (spec §3.3, ADR-0008)
bun run arch enforces: F1 pure domain · F2 application never imports adapters or cli · F3 contexts
only via index.ts; src/ and dashboard/ never import each other · F4 no cycles · F5 I/O (also
aliased globals, node:os, node:vm, node:module, bun:ffi, computed imports, eval) only in src/adapters, src/main.ts,
src/cli.ts, src/testing · F6 Agent SDK only in adapters/agent · F7 files ≤ 300 lines,
functions ≤ 40 · F8 tests use createTestEnv() · F9 no GitHub token or permission bypass for agents ·
F10 gh writes only in src/adapters/github/gh-github-writer.ts · F11 bash: no eval/sh -c/
python -c with expansions, no curl | sh, no ${{, no | grep -q, no BSD date -j/date -r · F12 dashboard layers, subprocesses only in
allowlisted services, no shell string from data · workflows never swallow gate or marker errors.
Bash and dashboard/server.ts debt is ratcheted (tools/arch/findings-baseline.json): never add a
finding; after removing some, run bun tools/arch/ast-scan.ts --write-baseline. Never suppress a rule
in place. Enforced later: F10 in full (FenceToken on run-scoped writes, Wave 1b; all integrations, Wave 2).
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.
- 2d ago Changed · +15 tokens per session 6ccf6d75c9ec
- 11d ago Changed · +13 tokens per session a5799a250711
- 12d ago Changed · +7 tokens per session 1b25538c0426
- 13d ago Changed · +6 lines · +228 tokens per session scan A → C 78023da898ee
- 14d ago First seen · 75 lines · 1,087 tokens per session scan A ca359f1fb09c
shipwright AGENTS.md is an instructions file published in the GitHub repository sethdford/shipwright (21 stars, last pushed today), licensed MIT. It adds 1,350 tokens to every session, about $0.0054 per session on Opus 5.5. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-15.
Other instructions, from other repositories
vscode buildNext.instructions.md
Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).
next.js AGENTS.md
AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.
vscode oss-third-party-notices.instructions.md
Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, cache eligibility and rollout and applying the notice (cutover).
langchain AGENTS.md
AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.
vscode copilot-instructions.md
Copilot instructions for microsoft/vscode, covering vs code copilot instructions, project overview, root folders, core architecture (src/ folder) and built-in extensions (extensions/ folder).
vscode model-prompts.instructions.md
Model-specific prompt authoring and registry guidelines.