MCP 助手
- 作者仓库星标 0
- 作者更新于 实时读取
- 作者仓库 skills-registry
- 领域
- AI 智能
- 兼容 Agent
-
- Claude Code
- Cursor
- Cline
- Codex
- Windsurf
- Gemini CLI
- +20
- 信任分
- 88 / 100 · 社区维护
- 作者 / 版本 / 许可
- @tomevault-io · 未声明 license
- Token 消耗评级
- 低消耗
- 接入复杂程度
- 需简单配置
- 是否需要外部 API Key
- 不需要
- 兼容的系统
- 未声明(默认跨平台)
- 底层运行要求
- 无特殊要求
- 文件与系统权限
-
- 只读
- 允许写入 / 修改
- Shell 执行
- 网络行为
- 允许外网请求
- 安装命令数
- 26 条
档案由构建时根据 SKILL.md 与安装命令自动衍生,可能与作者实际意图存在差异。
需要注意: 未限定 allowed-tools,默认拥有全部工具权限。
---
name: flutter-mcp-toolkit-custom-tools
description: Use this skill when the agent exposes app-specific surfaces by registering custom MCP tools and…
category: AI 智能
runtime: 无特殊运行时
---
# flutter-mcp-toolkit-custom-tools 输出预览
## PART A: 任务判断
- 适用问题:提示词、Agent 工作流、模型评估或自动化推理。
- 输入要求:目标材料、限制条件、期望输出和验收方式。
- 证据边界:围绕“Pick the right primitive / Handler signature (tools and resources) / Minimal tool registration”读取原文规则,不把推断写成作者承诺。
## PART B: 执行结果
- **01** 任务判断:确认你的需求是否属于提示词、Agent 工作流、模型评估或自动化推理,并标出输入、限制和预期结果。
- **02** 执行计划:优先按“Pick the right primitive / Handler signature (tools and resources) / Minimal tool registration”拆成步骤,说明每一步会读取什么、修改什么、产出什么。
- **03** 交付结果:给出可复制的命令、文件改动、检查清单或内容草稿,并说明如何继续迭代。
- **04** 风险边界:结合 读取文件、写入/修改文件、执行终端命令、会按任务需要访问外部网络、通常不需要额外 API Key 给出执行前确认项。
## Running Rules
- 读取文件、写入/修改文件、执行终端命令;会按任务需要访问外部网络;通常不需要额外 API Key。
- 先小样例验证,再放大到真实任务。
- 交付时同时给结果、检查口径和下一步迭代建议。 原文没有稳定的斜杠命令要求。安装验证后通常全局生效,直接在对话里点名这个 Skill 并描述任务即可。
告诉 Agent 目标文件或材料、期望结果、不可改范围、是否允许联网或执行命令。本 Skill 的权限画像是:读取文件、写入/修改文件、执行终端命令。
先用一个小任务确认它会围绕“Pick the right primitive / Handler signature (tools and resources) / Minimal tool registration”工作;涉及文件或命令时,先看 diff、日志、预览或测试结果。
检查最终产物是否包含明确结果、必要证据和下一步动作;如果输出泛泛而谈,就补充输入、边界和验收标准后重跑。
---
name: flutter-mcp-toolkit-custom-tools
description: Use this skill when the agent exposes app-specific surfaces by registering custom MCP tools and…
category: AI 智能
source: tomevault-io/skills-registry
---
# flutter-mcp-toolkit-custom-tools
## 什么时候使用
- 把 AI / Agent方向的常用动作沉淀成 Agent 可调用的技能 适合处理AI Agent、提示词、模型评估与自动化推理,核心价值是把输入、判断、执行、验证和交付边界固定下来,避免 Agent 泛泛回答。 把任务拆成可执行、可检查…
- 面向提示词、Agent 工作流、模型评估或自动化推理,优先处理能明确输入、步骤和验收标准的工作。
## 需要提供什么
- 目标材料、目录范围、期望结果和不可改动内容。
- 是否允许联网、执行命令、读写文件或调用外部服务。
## 执行规则
- 围绕「Pick the right primitive / Handler signature (tools and resources) / Minimal tool registration」组织步骤,不把推断写成作者事实。
- 读取文件、写入/修改文件、执行终端命令;会按任务需要访问外部网络;通常不需要额外 API Key。
- 先跑小样例,确认结果可检查后再扩大任务范围。
## 输出要求
- 给出最终产物、关键证据、验证方式和下一步动作。
- 信息不足时标记 unknown,不编造命令、平台或依赖。 作者原文负责流程事实;仓库文件负责来源和命令;流狐只补充适用场景、限制和质量判断。
skill "flutter-mcp-toolkit-custom-tools" {
输入层 -> 用户目标 + 目标文件 + 禁止范围 + 验收标准
上下文层 -> Pick the right primitive / Handler signature (tools and resources) / Minimal tool registration
规则层 -> SKILL.md 触发条件 / 执行顺序 / 输出格式
运行层 -> 无特殊运行时 | 读取文件、写入/修改文件、执行终端命令 | 会按任务需要访问外部网络
安全层 -> 通常不需要额外 API Key + 小任务验证 + diff / 日志复核
输出层 -> 可复制结果 + 检查清单 + 下一步迭代
} Custom MCP Toolkit Tools & Resources (Dynamic Registry)
Use this when bundled MCP tools (screenshot, semantic snapshot, tap, …) are not enough and you need app-specific read surfaces or actions — e.g. cart totals, feature flags, curated debug snapshots of internal state. Entries are registered in the Flutter process and exposed to the agent through the dynamic registry.
Pick the right primitive
| Need | Use |
|---|---|
| One-off read of a simple value | fmt_evaluate_dart_expression (no app code change). |
| Stable read-only payload (diagnostics, JSON snapshot, “current route”) | MCPCallEntry.resource + fmt_client_resource. Prefer resources when the contract is “GET-like” and idempotent. |
| Parameterized or mutating action, or reusable named operation | MCPCallEntry.tool + fmt_client_tool. |
Handler signature (tools and resources)
MCPCallHandler is FutureOr<MCPCallResult> Function(ServiceExtensionRequestMap request) where ServiceExtensionRequestMap is Map<String, String>.
- Tool arguments arrive as string values keyed by schema property names — mirror the README pattern:
request['n'],request['userId'], then parse (int.tryParse,double.tryParse,jsonDecodefor nested blobs if the wire format sends JSON-as-string). - Do not use
request.arguments— that is not the app-side API.
Minimal tool registration
import 'package:mcp_toolkit/mcp_toolkit.dart';
final tool = MCPCallEntry.tool(
handler: (request) async {
final userId = request['userId'] ?? '';
final cart = CartRepository.instance.forUser(userId);
return MCPCallResult(
message: 'ok',
parameters: {
'total': cart.total,
'items': cart.items.map((i) => i.toJson()).toList(),
},
);
},
definition: MCPToolDefinition(
name: 'cart_get_snapshot',
description: 'Return current cart total and items for a user.',
inputSchema: {
'type': 'object',
'additionalProperties': false,
'properties': {
'userId': {'type': 'string'},
},
'required': ['userId'],
},
),
);
await MCPToolkitBinding.instance.addEntries(entries: {tool});
Prefer MCPToolkitBinding.instance.bootstrapFlutter(additionalEntries: { ... }, runApp: ...) so tools/resources register in one place with zone/error setup — same entries shape as above.
Register after initialize() / bootstrapFlutter wiring, once at bootstrap — not inside build, not per-widget initState.
Custom resources
Resources are for read-only MCP surfaces: diagnostics, config summaries, or JSON blobs the agent polls without treating them as imperative actions.
MCPCallEntry.resource(
definition: MCPResourceDefinition(
name: 'app_cart_digest',
description: 'Compact cart summary for agents (read-only).',
mimeType: 'application/json',
),
handler: (request) async => MCPCallResult(
message: 'Cart digest',
parameters: {
'itemCount': CartRepository.instance.visibleCount,
'currency': CartRepository.instance.currencyCode,
},
),
),
namemust besnake_case(letters, digits, underscores).resourceUrimaps it to avisual://localhost/...URI (underscore segments become path segments). Agents consume it viafmt_client_resourceusing that URI / listing fromfmt_list_client_tools_and_resources.- Set
mimeTypehonestly (application/jsonvstext/plain) so clients know how to interpret payloads.
Schema rules (tools)
The MCP server enforces strict JSON Schema:
- Prefer
additionalProperties: falseunless you intentionally accept arbitrary keys. Unknown keys fail validation — good for catching agent typos. - Mark
requiredfor anything the handler reads unconditionally. - Prefer primitives and
enumover unconstrained strings. parametersinMCPCallResultmust be JSON-serializable; non-serializable objects degrade totoString().
Discovery from the agent side
fmt_list_client_tools_and_resources— enumerate app-registered tools and resources.fmt_client_tool— invoke a tool by name with JSON args (CLI:flutter-mcp-toolkit exec --name fmt_client_tool --args '...'per your transport).fmt_client_resource— fetch a registered resource (URI from listing /resourceUriconvention).
If something should appear but does not: confirm addEntries completed (await), then hot restart — reload does not always replay discovery cleanly.
Lifecycle gotchas
- Hot reload +
addEntriesfrom widget code → duplicate registrations. Register once inmain()/ bootstrap, not inbuild. - Hot restart clears VM state; registrations tied to
bootstrapFlutter/mainrun again on boot — correct pattern survives restart. - Debug mode only — release builds do not expose these VM service extensions.
- Naming: flat global namespace per app — prefix tools/resources (
cart_,flags_,nav_) to avoid collisions with builtins or other domains.
When the agent authors surfaces for the user’s app
- Ensure
mcp_toolkitis inpubspec.yaml. - Add
lib/mcp_tools/<domain>_surfaces.dartexportingregisterXSurfaces()that returnsSet<MCPCallEntry>or performsaddEntriesonce. - Wire
registerXSurfaces()frombootstrapFlutter(..., additionalEntries: ...)or calladdEntriesimmediately afterinitializeFlutterToolkitinsidebootstrapFlutter’s chain — never fromStatefulWidgetlifecycle. - Tight schemas (
additionalProperties: false, explicitrequired). - Hot restart, then
fmt_list_client_tools_and_resourcesbefore firstfmt_client_tool/fmt_client_resourcecall.
Safety and scope
- Treat handlers as powerful debug hooks: avoid exposing secrets, full databases, or unchecked filesystem/network IO.
- Keep handlers thin: delegate to domain/services already used by the app (same DI/getters), don’t duplicate business logic in MCP-only paths unless intentional.
Common traps
request.arguments— wrong shape; userequest['key']onMap<String, String>.- Missing
awaitonaddEntries→ race before discovery lists your surface. - Returning
Futureinstances insideparameters→ useless serialization;awaitinside the handler. inputSchemaout of sync with the handler → agents trust the schema; update both.
Related
- Driving the live app (snapshot / tap / reload):
flutter-mcp-toolkit-guide→flutter-mcp-toolkit-inspect/flutter-mcp-toolkit-control. - Repository
ARCHITECTURE.md→ “Dynamic Registry Architecture”.
Source: Arenukvern/mcp_flutter — distributed by TomeVault.
先判断是否适合
作者设计意图
作者的方法与取舍
边界和复核