// SCAN REPORT

Qflow

50 tools. 33 can modify or destroy data without limits.

1 destructive tool with no built-in limits. Policy required.

Last updated: 12 Jun 2026

33 can modify or destroy data

17 read-only

50 tools total

Community server · catalogue entry verified 12/06/2026

How to control Qflow ↓

TOOL MAP

What Qflow exposes to your agents

Read (17) Write / Execute (32) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Qflow tools

Destructive · High qflow_task_delete This tool irreversibly deletes tasks and their associated data. Execute · High qflow_plugin This tool manages plugins including installing and removing them. Execute · Medium qflow_tdd The tool runs TDD (Test-Driven Development) workflow actions including single-step execution and full loop cycles, which involve executing code or test processes. Execute · High qflow_autopilot This autopilot tool controls execution of automated processes with multiple operational commands (start, pause, resume, step, loop control).

33 of Qflow's 50 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Qflow

PolicyLayer is an MCP gateway — it sits between your AI agents and Qflow, and nothing reaches the server without passing your rules. These are the rules we recommend:

Deny destructive operations

{
  "qflow_task_delete": {
    "deny_if": [
      {
        "conditions": [],
        "on_deny": "Blocked by default. Requires approval."
      }
    ]
  }
}

Destructive tools should never be available to autonomous agents without human approval.

Rate limit write operations

{
  "qflow_parse_prd": {
    "limits": [
      {
        "counter": "qflow_parse_prd_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "qflow_onboarding": {
    "limits": [
      {
        "counter": "qflow_onboarding_per_minute",
        "window": "minute",
        "max": 60,
        "scope": "grant"
      }
    ]
  }
}

Controls API costs and prevents retry loops from exhausting upstream rate limits.

Create a free account and register Qflow — nothing to install.
Add these rules — paste them, or build them visually. Tune the limits to your setup.
Point your MCP client (Claude, Cursor, anything) at your gateway URL.

ENFORCE POLICY ON QFLOW →

Free to start. No card required.

FULL CATALOGUE

All 50 Qflow tools

DESTRUCTIVE 1 tools

Destructive qflow_task_delete 删除任务。cascade=true 时同时删除子任务，并清理其他任务中的依赖引用。

EXECUTE 5 tools

Execute qflow_plugin 插件管理：install=安装插件，remove=卸载，list=列表，get=详情，search=搜索，enable=启用，disable=禁用 Execute qflow_tdd 合并 TDD 工具。action: preset(查询预设)/step(单步执行)/loop(完整循环)/status(状态)/reset(重置)。 Execute qflow_autopilot 合并自动驾驶工具。action: config(配置)/start(启动)/pause(暂停)/resume(恢复)/status(状态)/stop(停止)/step(单步)/commit(提交)/loop_start( Execute qflow_workflow DAG 工作流管理：start=启动工作流，advance=推进工作流，status=查看状态，list=列出所有工作流 Execute qflow_agile 敏捷工作流预设：list=列出所有阶段，get=获取指定阶段，step=执行步骤

WRITE 27 tools

Write qflow_parse_prd 解析 PRD（产品需求文档）的 Markdown 内容，自动生成结构化任务列表。标题→任务，子标题→子任务。 Write qflow_scope_navigate 作用域导航：up=子任务提升为顶层任务, down=顶层任务降级为子任务, adjust=动态调节任务范围 Write qflow_constitution 合并治理原则工具。action: init(初始化)/get(获取列表)/set(添加原则)/validate(验证内容)。 Write qflow_context_compress 压缩上下文，按策略保留关键信息。aggressive 策略实际卸载模块释放 token，moderate 仅返回建议。 Write qflow_editor_rules 安装或查询编辑器工作规则。支持 13 种编辑器。install 安装规则文件（未指定 editor 时自动检测），list 列出支持的编辑器。 Write qflow_lottie Lottie 动画生成：从 34 种预置模板生成自定义 Lottie JSON 动画，支持颜色/尺寸/帧率自定义。 Write qflow_memory 合并记忆工具。action: flush(写入MEMORY.md)/load(读取MEMORY.md)。 Write qflow_plan_generate 从 Spec 生成实现计划（技术方案+数据模型+API契约+快速启动） Write qflow_profile_switch 切换到指定的 Profile 预设（自动应用 mode 和 contextModules）。 Write qflow_project_init 初始化项目的 qflow 配置。创建 .qflow/ 目录结构，包含配置文件和空任务列表。同时返回 onboardInfo 引导信息。 Write qflow_review 合并评审工具。action: create(创建评审)/comment(添加评论)/resolve(完成评审)/list(列出评审)/get(获取详情)/approve(创建审批)/reject(投票拒绝)/check( Write qflow_session_handoff 生成会话交接摘要（包含当前进度、待办事项、关键决策） Write qflow_spec_apply 应用待处理的 Spec 变更。确定性合并（RENAMED→REMOVED→MODIFIED→ADDED），原子写入。 Write qflow_spec_init 初始化一个新的 Spec 文档，创建 spec.json 和 spec.md。可通过 schemaId 关联自定义工作流 Schema。 Write qflow_spec_propose 提出 Spec 变更。默认 Delta 格式写入 changes/pending/。full=true 时一站式创建完整 Spec（含 proposal+design+tasks）。 Write qflow_spec_sync 将 Spec 内容同步到目标文件。deterministic 模式直接写入，agent 模式通过 AI 分析差异。 Write qflow_sprint 合并 Sprint 工具。action: create(创建)/get(获取详情)/add_story(添加故事)/update_story(更新故事状态)/complete(结束Sprint)。 Write qflow_tag_manage 标签管理：添加/移除/列表/过滤/重命名/删除/复制/从分支创建 Write qflow_task_batch 批量任务操作：创建/更新/查询状态/重写 Write qflow_task_create 创建新任务，自动评分复杂度。若复杂度 >= autoExpand 阈值，建议拆解子任务。 Write qflow_task_deps 依赖管理：验证/添加/移除/修复/可视化/关键路径分析 Write qflow_task_expand 将任务拆解为子任务（5 阶段模板），返回 hint 供宿主 LLM 进一步细化。 Write qflow_task_set_status 设置任务状态。当设为 done 时自动：解除阻塞的下游任务、推荐下一任务。返回更新结果和推荐的下一任务。 Write qflow_task_update 更新任务属性（标题、描述、优先级、依赖、标签等）。自动重新验证依赖图。 Write qflow_template 合并模板工具。action: create(创建模板)/apply(应用模板)/list(列出模板)。 Write qflow_use_tag 切换到指定的工作区标签。每个标签对应独立的任务文件，实现任务隔离。 Write qflow_workspace 合并工作区工具。action: isolate(隔离工作区)/switch(切换工作区)/merge(合并工作区)/status(工作区状态)。

READ 17 tools

Read qflow_onboarding 新手引导教程：init=初始化引导，step=获取当前步骤，complete=完成当前步骤，progress=查看进度，reset=重置引导，report=生成引导报告 Read qflow_clarification 需求澄清管理：ask=提出澄清问题，answer=回答问题，list=列出问题，unanswered=获取未回答问题 Read qflow_context_load 按需加载 CLAUDE.md 上下文模块。可用模块：core, phase1, phase2, ui-constraints, context-guard, thinking-tiers, iron-rules, rea Read qflow_context_status 获取已加载和可用上下文模块的状态，含逐模块 token 明细。 Read qflow_spec_status 获取所有 Spec 的状态概览，包括 Spec 总数、pending changes 和 applied changes 数量。 Read qflow_spec_verify 对 Spec 执行三维验证：完整性（content非空）、正确性（格式正确）、一致性（依赖存在）。 Read qflow_task_get 按 ID 获取任务详情，包含所有字段（标题、描述、状态、优先级、依赖、标签、子任务、元数据等）。 Read qflow_task_list 列出所有任务,支持按状态、标签过滤和模糊搜索。返回任务列表和统计信息。 Read qflow_tool_search 搜索已注册的 MCP 工具和斜杠命令。支持按名称/描述模糊匹配，按层级过滤。 Read qflow_complexity_score 对任务或描述进行复杂度评分（1-10），返回评分理由和拆解建议。 Read qflow_diagnostics 全系统健康检查：status=完整诊断，drift=配置漂移检测，watch_start=启动文件监控，watch_stop=停止监控，watch_events=获取监控事件。 Read qflow_report 项目报告：progress=进度报告（含状态统计、标签分组、阻塞分析），complexity=复杂度报告（含评分分布、高复杂度任务、拆解建议） Read qflow_research 聚合研究上下文数据（任务信息、文件内容、附加上下文），返回给宿主 LLM 进行分析。 Read qflow_spec_generate 从代码库生成 Spec。默认分析文件提取架构信息，structured=true 时 AI 生成结构化 Spec（含用户故事和验收条件）。 Read qflow_task_next 获取下一个推荐执行的任务。基于优先级、依赖状态和子任务关系自动选择最优任务。 Read qflow_task_tree 递归获取任务及其所有子任务的树结构 Read qflow_what_next 项目状态感知智能导航。自动检测项目阶段（init/planning/implementing/reviewing/done），推荐下一步操作。

EXPLORE

Related servers

Other MCP servers with similar tools — same risk classification, starter policies for each.

FAQ

Questions about Qflow

Can an AI agent delete data through the Qflow MCP server? +

Yes. The Qflow server exposes 1 destructive tools including qflow_task_delete. These permanently remove resources with no undo. PolicyLayer blocks destructive tools by default so they never reach the upstream server.

How do I prevent bulk modifications through Qflow? +

The Qflow server has 27 write tools including qflow_parse_prd, qflow_scope_navigate, qflow_constitution. Set a rate limit in your policy -- for example, 10 calls per hour prevents an agent from making more than 10 modifications per hour. PolicyLayer enforces this at the gateway, before calls reach Qflow.

How many tools does the Qflow MCP server expose? +

50 tools across 4 categories: Destructive, Execute, Read, Write. 17 are read-only. 33 can modify, create, or delete data.

How do I enforce a policy on Qflow? +

Register the Qflow MCP server in PolicyLayer, apply the suggested rules above (adjust the limits to your use case), and point your AI client at the PolicyLayer proxy URL instead of the server directly. Your agents keep the same tools; PolicyLayer evaluates every call against policy before it executes. Nothing to install, live in minutes.

Enforce policy on every Qflow tool call.

Deterministic rules across all 50 Qflow tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON QFLOW →

Free to start. No card required.

50 Qflow tools catalogued and risk-classified — across an index of 43,000+ MCP servers.