// SCAN REPORT

GitBranchMCP

34 tools. 17 can modify or destroy data without limits.

2 destructive tools with no built-in limits. Policy required.

Last updated: 29 Jun 2026

17 can modify or destroy data

17 read-only

34 tools total

Community server · catalogue entry verified 29/06/2026

How to control GitBranchMCP ↓

TOOL MAP

What GitBranchMCP exposes to your agents

Read (17) Write / Execute (15) Destructive / Financial (2)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous GitBranchMCP tools

Destructive · High delete_branch The delete_branch tool performs irreversible deletion of branch records/data, which is the defining characteristic of the Destructive category. Destructive · High reset_demo_data This tool irreversibly clears or overwrites demo data by resetting it to an initial state. Write · Low start_new_session The tool resets session state and returns to a clean mainline, which implies modifying/reinitializing session data rather than a destructive delete. Write · Medium answer_in_current_branch Based on the tool name and server purpose (branch notes and decision logging), this tool likely creates or modifies branch-level data by recording answers or messages.

17 of GitBranchMCP's 34 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control GitBranchMCP

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

Deny destructive operations

{
  "delete_branch": {
    "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

{
  "answer_in_current_branch": {
    "limits": [
      {
        "counter": "answer_in_current_branch_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "export_branch_markdown": {
    "limits": [
      {
        "counter": "export_branch_markdown_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 GitBranchMCP — 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 GITBRANCHMCP →

Instant setup, no code required.

FULL CATALOGUE

All 34 GitBranchMCP tools

DESTRUCTIVE 2 tools

Destructive delete_branch 当用户明确要求删除普通分支时调用。 Destructive reset_demo_data 当用户明确要求清空演示数据或恢复 demo 初始状态时调用。

WRITE 15 tools

Write answer_in_current_branch answer_in_current_branch Write append_message_node 手动向某个分支追加 message_node，用于验证 Git-like conversation graph。 Write branch_chat 在指定分支中进行隔离模型对话。 Write create_branch 当用户明确要求“开分支”“新建分支”“新探索分支”“另起支线”“单独研究一个问题”时调用。 Write create_branch_from_head create_branch_from_head Write create_demo_data 当用户想快速演示 Free Mode 能力或准备作品集展示时调用。 Write merge_branch_summary 将来源分支以 summary merge 方式合并到目标分支。 Write merge_summary_to_main 当用户要求把某个分支结论合并到主线时调用。 Write rename_branch 当用户明确要求修改普通分支名称时调用。 Write route_user_message route_user_message Write save_note 当分支讨论中出现重要结论、技术选择、方案取舍、需求变更、风险或待办时调用。 Write set_model_api_key 保存可选模型 API Key。 Write set_model_provider 设置可选模型 API provider 和 model。 Write start_new_session 当用户开始新的 Claude 聊天、要求重置当前 MCP 会话状态或想回到干净主线时调用。 Write switch_branch switch_branch

READ 17 tools

Read export_branch_markdown 当用户需要归档、交接或查看某个分支的 Markdown 记录时调用。 Read export_demo_markdown 当用户完成 demo 数据生成后，需要一次性导出演示 Markdown 文件时调用。 Read export_main_markdown 当用户需要导出主线决策记录、项目上下文摘要或交接材料时调用。 Read get_branch_context_preview 当用户要求预览当前/指定分支上下文、查看会发送给 API 模型的上下文时调用。 Read get_branch_diff 当用户要求查看某个 GitBranchMCP 分支自 fork 后新增内容、branch diff 时调用。 Read get_branch_log 当用户要求查看 GitBranchMCP 分支对话日志、branch log 或消息链时调用。 Read get_branch_notes 当用户要求查看某个 GitBranchMCP 分支笔记、分支结论或已保存记录时调用。 Read get_context_overview 当用户询问 GitBranchMCP 总览、当前上下文、有哪些分支、最近保存了什么时调用。 Read get_current_branch 当用户询问当前 GitBranchMCP 分支时调用。 Read get_merge_history 读取所有 summary merge 记录。 Read get_model_status 查看可选模型 API 配置状态。 Read get_workflow_prompt 当用户想配置 Claude Code、Cursor、Claude Desktop 项目规则或系统提示词时调用。 Read list_branches list_branches Read list_main_summaries 当用户想查看主线决策、已合并分支或历史摘要时调用。 Read suggest_branch_workflow 当用户不知道下一步如何使用分支工作流，或宿主模型需要给出固定操作建议时调用。 Read summarize_branch_diff 对来源分支 fork 后的 diff 生成结构化摘要。 Read test_model_connection 显式测试当前模型配置。

EXPLORE

Related servers

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

FAQ

Questions about GitBranchMCP

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

Yes. The GitBranchMCP server exposes 2 destructive tools including delete_branch, reset_demo_data. 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 GitBranchMCP? +

The GitBranchMCP server has 15 write tools including answer_in_current_branch, append_message_node, branch_chat. 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 GitBranchMCP.

How many tools does the GitBranch MCP server expose? +

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

How do I enforce a policy on GitBranchMCP? +

Register the GitBranch 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 GitBranchMCP tool call.

Deterministic rules across all 34 GitBranchMCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON GITBRANCHMCP →

Instant setup, no code required.

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