ultraship: Instructions file for Claude Code

CLAUDE.md

ultraship CLAUDE.md is an instructions file for Claude Code from Houseofmvps/ultraship. It costs 2,635 tokens per session, scanned A, original, MIT.

A project guide for the Ultraship plugin for Claude Code, an AI coding assistant. It describes the plugin’s folders, commands, specialist agents, skills, tools, and security-related parts.

In plain words
What is it for?
Use it when developing, reviewing, testing, securing, optimizing, or launching the Ultraship plugin.
Why use it?
It gives an agent a map of a large plugin so it can choose the right command, skill, or tool and follow the project’s conventions.

Instructions file for Claude Code

Written for Claude Code: ${CLAUDE_PLUGIN_ROOT} variable. Also seen: names the NotebookEdit tool; mentions Claude Code; mentions Codex.

This is Houseofmvps/ultraship's own configuration. It tells Claude Code how to work on ultraship itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything ultraship configures →

Runs only inside a plugin — its command needs a path that Claude Code sets for a plugin’s own hooks and for nothing else, and the catalogue could not identify which plugin ships it.

Reuse

Borrowing it

Nothing to install: this file belongs to Houseofmvps/ultraship. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/Houseofmvps/ultraship/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/Houseofmvps/ultraship

Made for: Claude Code.

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 ultraship CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/houseofmvps/ultraship/claude-md/github.svg)](https://agentmods.dev/instructions/houseofmvps/ultraship/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/houseofmvps/ultraship/claude-md"><img src="https://agentmods.dev/badge/instructions/houseofmvps/ultraship/claude-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.

agentmods 80×15 button for ultraship CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/houseofmvps/ultraship/claude-md"><img src="https://agentmods.dev/badge/instructions/houseofmvps/ultraship/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 2,635 This file is loaded in full into every session.
When invoked 2,635 The same file — it is already loaded in full.
Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.
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 $0.02635 $0.02635
Opus 5 $0.01318 $0.01318
Sonnet 5 $0.00527 $0.00527
Haiku 4.5 $0.00264 $0.00264

Measured 10d ago against content hash 522decb0cc2c, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

ultraship CLAUDE.md scanned grade A with 1 finding 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 10d 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.

Runs shell commandslowCapability

Expected in a hook, worth knowing in a rule or an instructions file.

- All tools use `execFileSync` (not `execSync`) to prevent shell injection
CLAUDE.md · 106 lines

How it starts

The opening of the file, as written. The whole thing — 106 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Ultraship — Claude Code Plugin

All-in-one builder plugin for Claude Code. npm: ultraship, GitHub: Houseofmvps/ultraship.

Project Structure

.claude-plugin/   — Plugin manifest (plugin.json)
skills/           — 45 skills (workflow + specialist + growth/launch/intelligence; includes staying-current, a11y, ship-gate, evals)
agents/           — 13 agents (review, seo, seo-strategist, security, pentest, perf, a11y, browser, compete, launch, incident, growth, canary)
commands/         — 16 command files (/ship, /seo, /secure, /review, /perf, /profile, /bundle, /deps, /health, /redirects, /codex, /content, etc.). NOTE: every skill is ALSO a slash command, so /a11y, /sprint, /pentest, /compete, /canary, /launch, /rescue, /grow etc. all work — the 21 redundant command launchers that duplicated same-named skills were removed 2026-06-28 (Anthropic merged commands into skills).
tools/            — 41 Node.js tools (ga4-client, keyword-intelligence, index-doctor, seo-scanner, a11y-scanner, ship-gate, vibe-security-scanner, eval-scanner, lighthouse, pentest-scanner, codex-generator, etc.) + shared lib/ship-scoring.mjs (single source of truth for /ship + ship-gate scoring)
hooks/            — SessionStart + UserPromptSubmit (Currency Guard) + PostCompact hooks + guard hooks (PreToolUse for destructive command blocking)
docs/             — Documentation

Tech Stack

  • Node.js ESM (type: module)
  • Zero external dependencies — HTML parsing via tools/lib/html-parser.mjs (inline SAX-style parser)
  • No build step — tools run directly via node tools/<tool>.mjs
  • All tools use execFileSync (not execSync) to prevent shell injection

Key Tools

Tool Purpose
seo-scanner.mjs SEO scanner (39 rules) + GEO signals (20 rules: AI bot access, snippet restrictions, structured data) + AEO signals (4 rules: schema checks). Cross-page analysis, analytics detection, canonical conflicts.
a11y-scanner.mjs Static accessibility scanner (WCAG 2.2 A/AA subset): missing alt, unlabeled form controls, empty buttons/links, missing lang/title/main, heading order, positive tabindex, zoom disabled, duplicate ids, broken aria references. Zero false positives. Dispatched by /ship; rendered contrast/focus checks escalate to npx pa11y via the /a11y skill.
ship-gate.mjs Blocking quality gate. Subcommands: init (write .ultraship/ship-gate.json thresholds), run (score via shared lib/ship-scoring.mjs, compare to thresholds, hard-fail on leaked secrets / critical findings, exit 1 on fail), ci (write .github/workflows/ship-gate.yml), hook (write .git/hooks/pre-push). Same scoring as /ship so they never disagree.
content-scorer.mjs Readability (Flesch-Kincaid), keyword density, GEO heading analysis
og-validator.mjs Open Graph tag validation, image reachability check
redirect-checker.mjs Redirect chain/loop detection, mixed protocol, sitemap-based bulk check
lighthouse-runner.mjs Lighthouse via headless Chrome, extracts CWV + diagnostics
health-check.mjs Production health check (status, response time, SSL, security headers)
env-validator.mjs Validates required env vars from .env.example against actual .env
migration-checker.mjs Detects pending DB migrations (Drizzle, Prisma, Knex)
bundle-tracker.mjs Bundle size tracking with history, heavy dependency detection
audit-history.mjs Saves/compares audit scores over time
api-smoke-test.mjs API endpoint smoke testing (status codes, response times, CORS)
gsc-client.mjs Google Search Console API (JWT auth via service account)
bing-webmaster.mjs Bing Webmaster API — 11 commands: submit-sitemap, list-sitemaps, submit-url, submit-url-batch, url-info, query, indexnow, keyword-research, backlinks, site-scan, url-inspection
code-profiler.mjs Static analysis for N+1 queries, sync I/O, memory leaks, unbounded queries
dep-doctor.mjs Detects unused/outdated dependencies, recommends removals
secret-scanner.mjs Detects leaked secrets, skips .env.example files
vibe-security-scanner.mjs Vibe-Coding Security Sentinel — context-aware checks secret-scanner misses: server-only secrets behind a public env prefix (NEXT_PUBLIC_/VITE_/etc.), a decoded Supabase service_role JWT exposed to the client, service_role referenced in a "use client" file, Supabase tables created without RLS (gated on a Supabase signal), and an advisory for mutation routes with no auth lib. Zero false positives.
eval-scanner.mjs Locates every LLM call site (Anthropic, OpenAI, Gemini, Mistral, Cohere, Ollama, Vercel AI SDK, LangChain) by provider + model id via real import/require detection (no local ./ai false positives), detects the test runner and whether a Promptfoo/eval suite already exists, and flags AI features shipping with no evals. Seeds the /evals skill. Zero false positives.
sitemap-generator.mjs Generates sitemap.xml from HTML files
robots-generator.mjs Generates AI-friendly robots.txt (allows GPTBot, PerplexityBot, etc.)
structured-data-generator.mjs Generates JSON-LD structured data
llms-txt-generator.mjs Generates llms.txt for AI discoverability
compete-analyzer.mjs Competitive X-Ray — compares two sites on tech stack, SEO, perf, security
launch-prep.mjs Launch Day Autopilot — generates launch copy, checklist, press kit
incident-commander.mjs Production Incident Commander — diagnoses outages, suggests rollback
growth-tracker.mjs Post-Ship Growth Intelligence — tracks metrics over time
cost-tracker.mjs AI Build Cost Tracker — tracks spend per feature/model
onboard-generator.mjs Instant Project Onboarding — generates developer onboarding guide
architecture-mapper.mjs Living Architecture Map — auto-generates Mermaid diagrams
pattern-analyzer.mjs Learn From the Best — analyzes and compares codebase patterns
demo-prep.mjs Demo-Ready Mode — finds dev artifacts, generates walkthrough
pentest-scanner.mjs Penetration testing — XSS, SQLi, SSTI, CORS, JWT, GraphQL, prototype pollution, race conditions. Zero false positives.
canary-monitor.mjs Post-deploy canary monitoring — HTTP status, response time, error patterns, baseline regression detection
retro-analyzer.mjs Sprint retrospective — git velocity, commit patterns, test health, hot files, shipping cadence
learnings-manager.mjs Project learnings CRUD — save, search, list, prune, export structured knowledge
ga4-client.mjs Google Analytics 4 Data API — 12 commands: overview, top-pages, traffic-sources, conversions, landing-pages, user-journey, search-terms, geo, devices, realtime, ai-traffic (ChatGPT/Perplexity/Copilot), organic. --organic flag for search-only filtering. Key event rates per page.
keyword-intelligence.mjs Keyword analysis — 12 commands: analyze (with difficulty + topical authority + SERP features + position bands), quick-wins, cannibalization, content-gaps, intent-map, trending, high-intent, page-keywords, content-decay, difficulty, anomalies (low-position high-CTR), cross-reference (GSC↔GA4). Brand filtering via --brand flag.
index-doctor.mjs Index diagnosis — inspects URLs via GSC, diagnoses non-indexing reasons (nosnippet/max-snippet detection), auto-submits fixes to Bing
codex-generator.mjs Codebase index — compact map of routes, schema, components, lib exports, structure. Stack-agnostic (JS/TS, Python, Go, Ruby, PHP, Rust, Java). Writes .ultraship/codex.md to save AI tokens. Skips audit fields + shadcn/radix UI primitives. Route paths validated (no false positives from non-route code).

Read the full file on GitHub · 106 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. 10d ago First seen · 106 lines · 2,635 tokens per session scan A 522decb0cc2c

Subscribe to this mod's changes

ultraship CLAUDE.md is an instructions file published in the GitHub repository Houseofmvps/ultraship (121 stars, last pushed 2mo ago), licensed MIT. It adds 2,635 tokens to every session, about $0.0132 per session on Opus 5. A static security scan graded it A with 1 finding (runs shell commands). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other instructions, from other repositories

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.

vercel/next.js · 7,296 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,153 tokens

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).

microsoft/vscode · 6,785 tokens

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

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.

langchain-ai/langchain · 4,469 tokens