Rails Ai Context logo

Rails Ai Context

已认领

作者:crisnahine

45 MCP tools that give AI coding agents ground truth about your Rails app: schema, models, routes, controllers, views, jobs, conventions. Works with Claude Code, Cursor, GitHub Copilot, OpenCode and Codex CLI. MCP or CLI, in-Gemfile or standalone, and it still answers when the app can't boot.

aiai-codingclaudecodexcodex-clicopilotcopilot-chatcopilot-clicursordeveloper-toolsintrospectionllmmcpmodel-context-protocolopencoderailsrails-airails-toolsrubyruby-gem

此服务器需要专门配置。目前没有可复用的公开启动命令,请遵循项目说明。

项目说明

更多Communication服务器

浏览完整目录

徽章

在 README 和网站展示此列表

徽章会随列表状态自动更新。认证后,列表会通过 do-follow 链接回链到你的网站,徽章与列表相互增强权威。展示徽章也是免费认证的前提。

安全概况

暴露面: 较广

会触及环境变量、凭据文件或其他代理的配置。

通过静态阅读发布代码得出的、此服务器能触及的范围。没有执行任何代码,这里的内容也不是判决:浏览器工具必须运行命令,API 封装必须调用它的 API。请自行判断是否适合你的环境。

NVIDIA SkillSpector v2.11.0,MCPVault MCP 服务器策略。扫描于今天 @ af2d257。

检测到

  • 读写其他代理的配置1 处

    触碰 ~/.claude、~/.codex 或 ~/.cursor 等目录,这些目录存放设置,有时也存放密钥。

  • 访问凭据文件13 处

    引用 SSH 密钥、云凭据文件或令牌存储。

  • 读取本地配置文件4 处

    加载 .env、.npmrc 或配置文件。对需要 API 密钥的服务器来说很正常。

  • 包含面向 AI 代理的指令4 处

    代码或面向代理的文件中有引导模型行为的文本。信任之前值得一读。

  • 未经检查地传递模型输出1 处

    模型输出未经验证就进入 HTML、SQL 或 shell。

未检测到: 批量读取环境变量, 向外发送对话内容, 运行 shell 命令, 使用提升的权限, 安装自身以便稍后运行, 匹配已知恶意模式.

带有已知公告的依赖

在与 OSV.dev 比对的锁文件和清单中未发现。

未纳入概况

文档、配置或 CI 文件中的 1 处模式匹配已记录,但不属于概况,因为服务器运行时它们不会执行。 拥有者仪表板会列出它们。

概况的工作原理 维护者可以在仪表板中看到每一项的文件、行号和修复建议。

40 个工具中显示 40 个

文档中列出的工具 (40)

内容来自项目文档。服务器握手不会验证每个工具的说明或行为。

analyze_feature

Full-stack: models + controllers + routes + services + jobs + views + tests

dependency_graph

Model/service dependency graph in Mermaid or text format

generate_test

Test scaffolding matching your project's patterns

get_active_support

Concerns registry, deprecators, MessageVerifier usage, onload hooks, cache store

get_autoload

Zeitwerk mode, autoload/eager-load paths, collapsed dirs, custom inflections

get_callbacks

Callbacks in Rails execution order with source

get_component_catalog

ViewComponent/Phlex: props, slots, previews, sidecar assets

get_concern

Concern methods + source + which models include it

get_config

Database, auth framework, assets, cache, queue, Action Cable

get_context

Composite: schema + model + controller + routes + views in one call

get_controllers

Actions + inherited filters + render map + strong params

get_conventions

Auth checks, flash messages, create action template, test patterns

get_edit_context

Method-aware code extraction with class context

get_engines

Mounted engines (with known-engine descriptions) + loaded engine classes

get_env

Environment variables + credentials keys (not values)

get_env_config

Per-environment config: notable toggles + every config key each env sets

get_frontend_stack

React/Vue/Svelte/Angular, Hotwire, TypeScript, package manager

get_gems

Notable gems with versions, categories, config file locations

get_helper_methods

App + framework helpers with view cross-references

get_i18n

Locales, translation files with key counts, per-locale coverage, fallbacks

get_job_pattern

Queue, retries, guard clauses, broadcasts, schedules

get_mailers

Mailer classes with delivery actions and delivery method

get_model_details

AST-parsed associations, validations, scopes, enums, macros - each result tagged [VERIFIED] or [INFERRED]

get_partial_interface

What locals to pass + what methods are called on them

get_routes

Code-ready helpers (postpath(@record)) + required params

get_schema

Columns with indexed/unique/encrypted/default hints

get_service_pattern

Interface, dependencies, side effects, callers

get_stimulus

HTML data-attributes (dashes!) + targets + values + actions

get_test_info

Fixtures + relationships + test template matching your patterns

get_turbo_map

Broadcast → subscription wiring + mismatch warnings

get_view

Templates with ivars, Turbo wiring, Stimulus refs, partial locals

migration_advisor

Migration code generation with reversibility + affected models

onboard

Narrative app walkthrough (quick/standard/full)

performance_check

N+1 risks, missing indexes, countercache, eager load candidates

query

Safe read-only SQL with timeout, row limit, column redaction

read_logs

Reverse file tail with level filtering and sensitive data redaction

search_code

Trace: definition + source + callers + tests. Also: definition, call, class filters

search_docs

Bundled topic index with weighted keyword search

security_scan

Brakeman static analysis - SQL injection, XSS, mass assignment

validate

Syntax + semantic + Brakeman security in one call

工具变更历史

比较相同配置的完整检查。只列出工具,未调用工具。此处不测量输入结构的变化。

尚无完整工具检查。