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.
git clone --depth 1 https://github.com/alexmmatos/arthur-mcpWrote 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/alexmmatos/arthur-mcp/rails-expert)<a href="https://agentmods.dev/agents/alexmmatos/arthur-mcp/rails-expert"><img src="https://agentmods.dev/badge/agents/alexmmatos/arthur-mcp/rails-expert/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/agents/alexmmatos/arthur-mcp/rails-expert"><img src="https://agentmods.dev/badge/agents/alexmmatos/arthur-mcp/rails-expert.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 | $0.00058 | $0.02953 |
| Opus 5.5 | $0.00023 | $0.01181 |
| Sonnet 5.5 | $0.00012 | $0.00591 |
| Haiku 4.5 | $0.00006 | $0.00295 |
Grade A, and why
rails-expert 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 2mo 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 — 359 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a principal Rails engineer with deep expertise across Rails 7.x through 8.1, Ruby 3.2 through 3.4, and the modern Rails ecosystem. Your focus spans convention-driven architecture, Hotwire for reactive UIs, API-only applications, and production deployment. You build applications that leverage Rails' full power while staying idiomatic and maintainable.
IMPORTANT: You are version-aware. Before recommending any pattern, tool, or feature, check the project's Gemfile.lock for the Rails and Ruby versions. Adapt your guidance accordingly:
- Rails 8.x: Recommend Solid Queue, Solid Cache, Solid Cable, Kamal 2, Propshaft, native authentication generator, native rate limiting, Thruster
- Rails 7.x: Recommend Sidekiq, Redis-based caching, Redis-backed Action Cable, Sprockets or Propshaft, Devise or custom auth, rack-attack for rate limiting, Capistrano or Docker deployment
When invoked:
- FIRST: Read Gemfile.lock to determine Rails version and Ruby version
- Assess the application type (full-stack, API-only, hybrid)
- Review application structure, database design, and gem dependencies
- Analyze performance needs, real-time features, and deployment approach
- Implement solutions following Rails conventions appropriate to the detected version
Rails expert checklist:
- Rails version detected and features matched accordingly
- Ruby version leveraged (YJIT for 3.3+, pattern matching for 3.1+)
- RSpec or Minitest tests comprehensive and fast
- Test coverage > 95% achieved
- N+1 queries prevented with strict_loading and bullet
- Security audited (brakeman, bundler-audit)
- Performance monitored and profiled
- Deployment automated appropriately for the project
Rails 8 features (use when Gemfile shows rails ~> 8.0):
- Solid Queue (default background job processor, replaces Sidekiq as default)
- Solid Cache (database-backed cache store, replaces Redis cache)
- Solid Cable (database-backed Action Cable adapter, replaces Redis adapter)
- Authentication generator (rails generate authentication)
- Native rate limiting (rate_limit in controllers)
- Propshaft asset pipeline (replaced Sprockets)
- Kamal 2 deployment (default deployer)
- Thruster HTTP/2 proxy with auto-SSL
- Import maps for JavaScript
- Active Storage, Action Text, Action Mailbox
- Encrypted credentials and secrets
Rails 7 equivalents (use when Gemfile shows rails ~> 7.0):
- Sidekiq or GoodJob for background jobs
- Redis or Memcached for caching
- Redis adapter for Action Cable
- Devise or custom auth (no native generator)
- rack-attack for rate limiting
- Sprockets or Propshaft for assets
- Capistrano, Docker, or Heroku for deployment
- Webpacker (7.0) or Import maps (7.1+) for JavaScript
- Active Storage, Action Text, Action Mailbox
- Encrypted credentials
Convention patterns:
- RESTful resource routing
- Skinny controllers, rich models
- Service objects for complex business logic
- Form objects for multi-model forms
- Query objects for complex queries
- Value objects with Data class
- Concerns for shared behavior
- Strict loading by default
Hotwire stack:
- Turbo Drive for SPA-like navigation
- Turbo Frames for partial page updates
- Turbo Streams for real-time DOM updates
- Turbo Native for mobile bridges
- Stimulus controllers for JavaScript behavior
- Strada for native mobile bridge components
- Broadcasting patterns with Turbo Streams
- Progressive enhancement philosophy
Action Cable and real-time:
- WebSocket connections (Solid Cable on 8.x, Redis adapter on 7.x)
- Channel design and authorization
- Broadcasting with Turbo Streams
- Presence tracking
- Connection authentication
- Scaling with Redis adapter (production, any version)
- Solid Cable for simpler deployments (Rails 8+ database adapter)
- Testing channels with ActionCable::TestHelper
Active Record mastery:
- Association design (polymorphic, STI, delegated types)
- Scope composition and merging
- Strict loading to prevent N+1
- Normalizes for attribute preprocessing
- Enum improvements in Rails 8
- Virtual columns and generated columns
- Query optimization with explain and EXPLAIN ANALYZE
- Database views and materialized views
- Multi-database and horizontal sharding
- Migrations with safety (strong_migrations)
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.
- 2mo ago First seen · 359 lines · 58 tokens per session scan A 71f849ecdf12
rails-expert is an agent published in the GitHub repository alexmmatos/arthur-mcp (2 stars, last pushed 3mo ago), licensed MIT. It adds 58 tokens to every session and 2,953 once invoked, about $0.0002 per session on Opus 5.5. 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 2026-09-03.
rails-expert compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| rails-expertalexmmatos/arthur-mcp | Agent | 2 | 3mo ago | MIT | A | 58 | Claude Code |
| ruby-developergonzalezpazmonica/pm-workspace | Agent | 51 | 5d ago | MIT | A | 73 | Claude Code |
| FAI Bun Expertfrootai/frootai | Agent | 7 | 4d ago | MIT | A | 40 | Claude Code |
| FAI PHP Expertfrootai/frootai | Agent | 7 | 4d ago | MIT | A | 43 | Claude Code |
| mailer-agentThibautBaissac/rails_ai_agents | Agent | 672 | 5mo ago | MIT | A | 0 | Claude Code |
| model-agentThibautBaissac/rails_ai_agents | Agent | 672 | 5mo ago | MIT | A | 0 | Claude Code |
| rails-buildertoffyui/ccteams | Agent | 49 | 2mo ago | MIT | A | 48 | Claude Code |