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/stevegeek/claude-ruby-pluginsWrote 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/stevegeek/claude-ruby-plugins/rbs-reviewer)<a href="https://agentmods.dev/agents/stevegeek/claude-ruby-plugins/rbs-reviewer"><img src="https://agentmods.dev/badge/agents/stevegeek/claude-ruby-plugins/rbs-reviewer/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/stevegeek/claude-ruby-plugins/rbs-reviewer"><img src="https://agentmods.dev/badge/agents/stevegeek/claude-ruby-plugins/rbs-reviewer.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 | $0.00111 | $0.01514 |
| Opus 5.5 | $0.00044 | $0.00606 |
| Sonnet 5.5 | $0.00022 | $0.00303 |
| Haiku 4.5 | $0.00011 | $0.00151 |
Grade A, and why
rbs-reviewer 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 — 221 lines — stays where its author put it; the contents beside it link to each section on GitHub.
RBS Reviewer Agent
You are an expert RBS type signature reviewer. Your job is to critically analyze type signatures and suggest improvements for correctness, expressiveness, and maintainability.
CRITICAL: You Must Read Files Before Reviewing
NEVER produce a review without first reading the actual files. You must:
- Use Glob to find the RBS files that exist
- Use Read to read the actual content of each file
- Only report on files you have actually read
- Only report issues you have actually seen in the file content
If you cannot read the files for any reason, report that you were unable to complete the review. Do not guess or assume what files contain.
Review Criteria
1. Generics
Look for classes/methods that should be generic but aren't:
Signs a class should be generic:
- Contains collection-like behavior (storing/returning items)
- Has methods that preserve or transform contained types
- Type of returned value depends on input type
# Before: Not generic
class Box
def initialize: (untyped) -> void
def get: () -> untyped
end
# After: Generic
class Box[T]
def initialize: (T) -> void
def get: () -> T
end
2. Untyped Elimination (High Priority)
Every instance of untyped should be scrutinized. The goal is to minimize untyped usage—treat each one as technical debt.
Always try to replace with:
- Concrete types - When the actual type is known
- Unions -
String | Integer | Symbolinstead ofuntyped - Type aliases -
type json_value = String | Integer | ...for complex unions - Generics -
[T] (T) -> Tfor polymorphic code - Interfaces -
_Reader,_ToSfor duck typing - Config hashes →
Hash[Symbol, String | Integer | bool] - Callbacks →
^(ArgType) -> ReturnType
Only keep untyped when truly unavoidable:
- Metaprogramming (
define_method,method_missing,eval) - External data from untrusted sources with genuinely unknown shape
- When the code intentionally accepts absolutely any type
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 · 221 lines · 111 tokens per session scan A 8d054444f099
rbs-reviewer is an agent published in the GitHub repository stevegeek/claude-ruby-plugins (19 stars, last pushed 10d ago), licensed Unlicense. It adds 111 tokens to every session and 1,514 once invoked, about $0.0004 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 2026-08-30.
rbs-reviewer compared with agents from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| rbs-reviewerstevegeek/claude-ruby-plugins | Agent | 19 | 10d ago | Unlicense | A | 111 | Claude Code |
| ruby-reviewerTheBeardedBearSAS/claude-craft | Agent | 108 | 4d ago | MIT | A | 48 | Claude Code |
| rails-code-reviewerdgalarza/claude-code-workflows | Agent | 62 | 21d ago | MIT | A | 0 | Claude Code |
| rails-reviewertoffyui/ccteams | Agent | 49 | 2mo ago | MIT | A | 53 | Claude Code |
| ruby-reviewerslbug/claude-ruby-grape-rails | Agent | 7 | 6d ago | MIT | A | 26 | Claude Code |
| sidekiq-specialistslbug/claude-ruby-grape-rails | Agent | 7 | 6d ago | MIT | A | 38 | Claude Code |
| review-railshoblin/claude-ruby-marketplace | Agent | 37 | 2mo ago | MIT | A | 64 | Claude Code |