// SCAN REPORT

Booster MCP

31 tools. 7 can modify or destroy data without limits.

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

Last updated: 3 Jul 2026

7 can modify or destroy data

24 read-only

31 tools total

Community server · catalogue entry verified 03/07/2026

How to control Booster MCP ↓

TOOL MAP

What Booster MCP exposes to your agents

Read (24) Write / Execute (6) Destructive / Financial (1)

Running more than one server? Check your whole stack →

MOST DANGEROUS TOOLS

Critical Risk

The most dangerous Booster MCP tools

Destructive · High remove_repo This tool permanently removes a repository from indexing, which is an irreversible operation that cannot be undone without re-adding and re-indexing the repository. Execute · Critical run_command Running shell commands is a classic Execute action with critical severity due to unrestricted blast radius—an AI agent could execute any command (data exfiltration, system modification, lateral movement, credential theft Write · Medium add_repo This tool creates a new indexed entry for a repository, which is a reversible write operation. Write · Low flipchart_add_note The tool creates or appends data (a note/insight) to an existing flipchart session.

7 of Booster MCP's 31 tools can modify, destroy, or commit something on every call — and an agent calls them with no built-in limits.

POLICY

How to control Booster MCP

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

Deny destructive operations

{
  "remove_repo": {
    "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

{
  "add_repo": {
    "limits": [
      {
        "counter": "add_repo_per_hour",
        "window": "hour",
        "max": 30,
        "scope": "grant"
      }
    ]
  }
}

Prevents bulk unintended modifications from agents caught in loops.

Cap read operations

{
  "analyze_error": {
    "limits": [
      {
        "counter": "analyze_error_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 Booster MCP — 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 BOOSTER →

Instant setup, no code required.

FULL CATALOGUE

All 31 Booster MCP tools

DESTRUCTIVE 1 tools

Destructive remove_repo Удаляет репозиторий из списка индексации (данные сохраняются в индексе).

EXECUTE 1 tools

Execute run_command Выполняет команду в shell. timeout в миллисекундах

WRITE 5 tools

Write add_repo Добавляет репозиторий для индексации и индексирует его. Write flipchart_add_note Добавляет заметку-инсайт на флипчарт сессии. Write flipchart_create_session Создаёт сессию дебага для отслеживания группы символов. Write install_agent_skills Синхронизирует встроенные agent skills в ~/.agents/skills. Write project_memory Управление долгосрочной памятью проекта.

READ 24 tools

Read analyze_error Ищет в коде потенциальные причины ошибки по тексту stacktrace Read code_grep Регулярный поиск по всем файлам проектов Read compare_symbols Сравнивает реализацию символа в двух файлах через diff Read external_deps Находит внешние зависимости: HTTP, БД, Redis, Kafka, файлы, subprocess Read fetch_stack_docs Анализирует технологический стек проекта и возвращает рекомендации агенту: Read find_duplicates Ищет дублирующиеся блоки кода (copy-paste) Read find_symbol Ищет функцию или класс по имени. Read flipchart_call_graph Генерирует Mermaid-диаграмму вызовов для символа. Read flipchart_get_board Возвращает полный флипчарт сессии: все диаграммы и заметки. Read flipchart_quick_debug Быстрый дебаг символа: генерирует Mermaid call graph + семантический контекст. Read flipchart_sequence_diagram Генерирует Sequence-диаграмму выполнения от символа. Read get_code_city get_code_city Read get_repo_map Генерирует сжатую карту репозитория в стиле Aider RepoMap. Read git_diff Показывает изменения в файле/репозитории Read git_log История коммитов для файла/репозитория Read inject_context Собирает полный контекст по проекту (repo map, стек и конвенции), Read list_agent_skills Возвращает список встроенных agent skills, поставляемых вместе с MCP. Read list_configs Находит все конфигурационные файлы в репозитории Read list_repos Возвращает список всех репозиториев под управлением MCP. Read read_file Читает диапазон строк из файла (0-indexed) Read read_with_context Читает файл с ±N строк вокруг указанной линии Read reindex_repo Переиндексирует указанный репозиторий (полная очистка и новая индексация). Read repo_stats Возвращает статистику проиндексированного репозитория. Read semantic_search Ищет фрагменты кода по смыслу (векторный поиск).

EXPLORE

Related servers

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

FAQ

Questions about Booster MCP

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

Yes. The Booster MCP server exposes 1 destructive tools including remove_repo. 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 Booster MCP? +

The Booster MCP server has 5 write tools including add_repo, flipchart_add_note, flipchart_create_session. 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 Booster MCP.

How many tools does the Booster MCP server expose? +

31 tools across 4 categories: Destructive, Execute, Read, Write. 24 are read-only. 7 can modify, create, or delete data.

How do I enforce a policy on Booster MCP? +

Register the Booster 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 Booster MCP tool call.

Deterministic rules across all 31 Booster MCP tools. Per-identity grants. Full audit log. Live in minutes. Nothing to install.

ENFORCE POLICY ON BOOSTER →

Instant setup, no code required.

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