Сокращает до 98% токенов, которые ваши AI-агенты тратят на использование инструментов MCP — без потерь.
Сервер | Инструментов | До | После | Сокращение |
filesystem (Anthropic) | 14 | 1 892 ток | 236 ток | −88% |
sequential-thinking | 1 | 926 ток | 20 ток | −98% |
Google Calendar | 5 | 5 010 ток | 660 ток | −87% |
Enterprise (Cal + Gmail + Drive) | 12 | 8 649 ток | 882 ток | −90% |
sample_app.js (JavaScript) | — | 799 ток | 284 ток | −64.5% |
sample_app.ts (TypeScript) | — | 378 ток | 266 ток | −29.6% |
ast_extractor.py (Python) | — | 3 633 ток | 890 ток | −75.5% |
Меньше отправленных токенов = ниже счета за API и быстрее ответы. И ничего не теряется. Каждая проверка подтверждает, что инструменты остаются на 100% работоспособными после сжатия.
Воспроизведите эти числа самостоятельно — каждый входной файл статический фикстур в репозитории, токены подсчитаны с помощью tiktoken cl100k_base:
python benchmarks/run_benchmark.py # the table above
python benchmarks/run_benchmark.py --json # machine-readable
python benchmarks/run_benchmark.py --fixture path/to/your_schemas.jsonRelated MCP server: Bifrost-MCP Gateway
Одной строкой (macOS / Linux) — устанавливает пакет и настраивает Claude Desktop:
curl -sSL https://raw.githubusercontent.com/LoudiliMed/Refract/main/install-refract.sh | shПередача скрипта из интернета в оболочку требует беглого просмотра: посмотреть install-refract.sh на GitHub. Он никогда не использует sudo.
Или напрямую через pip:
pip install refract-mcpДополнительные возможности:
pip install refract-mcp[semantic] # semantic tool routing with embeddings
pip install refract-mcp[multilang] # JavaScript, TypeScript, JSX, TSX supportРасполагается между вашим агентом и любым MCP-сервером. Сжимает схемы инструментов на лету, чтобы агент не загружал полный каталог при каждом запросе.
Локальный подпроцесс (stdio):
refract-proxy --target "npx @modelcontextprotocol/server-filesystem /tmp" --verboseУдалённый HTTP/SSE-сервер:
# --url implies SSE transport (explicit, recommended for remote endpoints)
refract-proxy --url https://my-mcp-server.com/sse
# or with --transport flag (auto-detection can be overridden)
refract-proxy --target https://my-mcp-server.com/sse --transport sseФлаги прокси:
Флаг | По умолчанию | Описание |
| обязательный | Цель MCP: команда stdio, HTTP URL или JSON-файл |
| — | Псевдоним для |
| — | Удалённая конечная точка SSE/HTTP — подразумевает |
| auto | Принудительный транспорт к цели: |
| 30 | Тайм-аут подключения для SSE-целей (3 повтора) |
| stdio | Как прокси обслуживает вашего агента |
| 8080 | Порт прослушивания прокси в режиме |
| выкл | Выводить количество токенов на каждый запрос |
| WARNING | DEBUG / INFO / WARNING / ERROR |
Добавление в Claude Desktop:
{
"mcpServers": {
"my-server-via-refract": {
"command": "/path/to/refract-proxy",
"args": [
"--target",
"npx @modelcontextprotocol/server-filesystem /path/to/folder",
"--verbose"
]
}
}
}Для удалённого MCP-сервера (SSE):
{
"mcpServers": {
"remote-via-refract": {
"command": "/path/to/refract-proxy",
"args": ["--url", "https://my-mcp-server.com/sse"]
}
}
}Вместо того чтобы редактировать записи по одной (или запускать refract-install для каждого сервера), refract-wrap-all переписывает все stdio-серверы в claude_desktop_config.json так, чтобы они проходили через refract-proxy одной командой:
# Preview what would change — writes nothing
refract-wrap-all --dry-run
# Wrap every stdio server not already going through refract
refract-wrap-all
# Restore the original commands
refract-wrap-all --unwrapПример: эта запись
{
"mcpServers": {
"filesystem": {
"command": "npx",
"args": ["@modelcontextprotocol/server-filesystem", "/tmp"],
"env": {"MY_VAR": "1"}
}
}
}превращается в
{
"mcpServers": {
"filesystem": {
"command": "/path/to/refract-proxy",
"args": ["--stdio-cmd", "npx @modelcontextprotocol/server-filesystem /tmp"],
"env": {"MY_VAR": "1"}
}
}
}Гарантии:
Резервная копия конфигурации создаётся перед каждой записью (claude_desktop_config.json.bak, затем .bak2, .bak3… — существующая резервная копия никогда не перезаписывается).
Поля env, cwd и любые другие поля сервера сохраняются.
Серверы, уже проходящие через refract-proxy или refract-server, пропускаются (already wrapped).
Удалённые SSE/HTTP-серверы (записи с url) пропускаются — обёртываются только stdio-серверы.
--unwrap — точная обратимая операция: wrap затем unwrap восстанавливает исходную команду, аргументы и поля.
Флаг | Значение | Описание |
| Streamable HTTP | Текущий стандарт (спецификация MCP 2025-03-26). Используйте с удалёнными MCP-серверами. |
| SSE | Устаревший транспорт, оставлен для совместимости. Используйте, если сервер не поддерживает Streamable HTTP. |
| stdio-подпроцесс | Локальная команда (по умолчанию, когда |
(опущен) | автоопределение | Определяется по |
Оба sse и http требуют HTTP(S) URL в --target.
# Connect to a remote MCP server via Streamable HTTP (recommended)
refract-proxy --target "https://my-mcp-server.com/mcp" --transport http
# Connect via SSE (legacy)
refract-proxy --target "https://my-mcp-server.com/sse" --transport sse
# Local subprocess (auto-detected, --transport stdio optional)
refract-proxy --target "npx @modelcontextprotocol/server-filesystem /tmp"Предоставляет вашу кодовую базу как MCP-сервер. Ваш агент может проиндексировать репозиторий, получить сжатый контекст файла, развернуть конкретные функции, проанализировать влияние, обнаружить критические изменения и нанести на карту риски безопасности.
refract-server --root /path/to/your/repoДобавление в Claude Desktop:
{
"mcpServers": {
"refract-code": {
"command": "/path/to/refract-server",
"args": ["--root", "/path/to/your/repo"]
}
}
}Представьте библиотеку с 50 книгами.
Без Refract: ваш агент получает подробное описание всех 50 книг на каждый вопрос, даже если ответу нужна только одна.
С Refract: агент сначала получает список названий (индекс). Как только он понимает, какая книга нужна, он получает только содержимое этой книги.
Технически:
Индекс (всегда отправляется): только названия инструментов и краткое описание каждого.
Детали (отправляются только при необходимости): полное описание используемого инструмента, всё необходимое для его правильного использования, ничего лишнего.
Проверка: после каждого сжатия Refract автоматически проверяет, что ничего важного не удалено. Если есть сомнения, отправляется полная версия — без риска.
В этом процессе не задействована ни одна AI-модель. Он полностью автоматический, быстрый и детерминированный.
Инструмент | Что делает |
Сжатие | Сжимает схемы инструментов на лету, до 98% сокращения |
Проверка сигнала | Проверяет вызываемый контракт после каждого сжатия |
Семантическая маршрутизация | Определяет нужный инструмент с помощью эмбеддингов (опционально) |
Кэширование подсказок | Вставляет Anthropic cache_control для повторяющихся запросов |
Инструмент | Вход | Выход |
index_repo | путь к репозиторию | агрегированный индекс всех файлов Python, JS, TS |
get_compressed | путь к файлу | сжатая структура + статистика токенов |
expand | путь к файлу + имена функций | исходный код + контекст зависимостей |
blast_radius | путь к файлу + имя функции | все функции, которые сломаются при изменении цели |
semantic_diff | путь к файлу + старый исходник + новый исходник | критические изменения против изменений только тела |
semantic_diff_branches | путь к репозиторию + файл + функция + базовая/целевая ссылки git | semantic_diff одной функции между двумя ветками/коммитами |
security_surface | путь к репозиторию | карта опасных вызовов (subprocess, eval, pickle, requests) |
refract-status --root /path/to/repo
refract-status --root /path/to/repo --jsonФлаг | Описание |
| Путь для анализа (по умолчанию: текущая директория) |
| Машиночитаемый вывод |
Показывает: файлы по языкам, исходные и сжатые токены, функции/классы в индексе, опасные вызовы по категориям, языки без поддержки tree-sitter.
Попросите Claude показать, какие функции сломаются, если вы измените целевую функцию.
Пример результата:
{
"target": "authenticate",
"direct_callers": ["login_user"],
"all_impacted": ["login_user", "verify_session", "admin_access"],
"impacted_count": 3,
"risk_level": "MEDIUM"
}Уровни риска: LOW (от 0 до 2 затронутых), MEDIUM (от 3 до 5), HIGH (6 и более).
Обнаруживает критические изменения API, сравнивая интерфейсы функций, а не их тела. Используйте как шлюз CI.
Пример результата:
{
"breaking": ["authenticate"],
"body_only": ["logout"],
"added": ["new_function"],
"removed": [],
"unchanged": ["hash_password"],
"is_breaking": true
}Если is_breaking равно true, PR изменяет публичный API и требует проверки.
Наносит на карту каждую функцию, вызывающую опасные примитивы, во всём репозитории.
HIGH риск: subprocess, os.system, eval, exec, pickle, ctypes
MEDIUM риск: open (режим записи), socket, requests, httpx, urllib
Пример результата:
{
"high_risk": [
{
"file": "src/llm_client.py",
"function": "run_command",
"calls": ["subprocess.run"]
}
],
"summary": {
"high_risk_count": 1,
"medium_risk_count": 3,
"total_functions_scanned": 87,
"clean_files": 8
}
}Python (через модуль ast), JavaScript, TypeScript, JSX, TSX (через tree-sitter, опционально: pip install refract-mcp[multilang]).
Язык определяется автоматически по расширению файла. Элегантный откат, если tree-sitter не установлен.
Refract интегрируется с кэшированием подсказок Anthropic. as_anthropic_tools() автоматически помечает сжатый каталог как кэшируемый, что ещё больше снижает затраты на повторные запросы.
Пример за 30 дней, 100 запросов в день, 5 000 токенов схем:
Сценарий | Стоимость |
Без Refract, без кэша | $45.00 |
С Refract + кэш | $1.49 |
"Failed to spawn process: No such file or directory" в Claude Desktop
Claude Desktop не может найти refract-proxy в своём PATH. Найдите абсолютный путь и используйте его напрямую:
which refract-proxyЗатем используйте полный путь в claude_desktop_config.json:
{
"mcpServers": {
"my-tool-via-refract": {
"command": "/full/path/to/refract-proxy",
"args": [
"--target",
"npx @modelcontextprotocol/server-filesystem /path/to/folder"
]
}
}
}Claude Desktop, Cursor, любым клиентом, соответствующим стандарту MCP, любым существующим MCP-сервером.
MIT — свободное использование, включая коммерческое.
Reverse-call-graph impact analysis for a Python function: BFS over inverted call edges to find every function transitively affected by changing target_function. Returns direct_callers, all_impacted, impacted_count, total_functions and a risk_level. Python files only.
| Name | Required | Description | Default |
|---|---|---|---|
| file_path | Yes | Path to a .py file (relative to --root or absolute). | |
| target_function | Yes | Function/method name to analyze the blast radius of. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description fully discloses behavior: BFS algorithm, inverted call edges, returns specific fields (direct_callers, all_impacted, etc.), and scope (Python files only). No hidden side effects or destructive actions are implied.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two sentences, front-loads the purpose, and includes all essential information without verbosity. Every sentence earns its place.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given no annotations or output schema, the description covers purpose, algorithm, input parameters, return fields, and file type restriction. It is complete for an agent to understand and use this tool correctly.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% (both params described), so baseline is 3. The description adds value by explaining the algorithm context (BFS over inverted call edges) and linking the parameters to the analysis process, though the schema already defines them clearly.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states 'Reverse-call-graph impact analysis for a Python function', specifying the verb (analyze impact), resource (Python function), and methodology (BFS over inverted call edges). It distinguishes from sibling tools like index_repo and security_surface which have different purposes.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implicitly defines usage: when you need to know which functions are affected by changing a target function in a Python file. It does not explicitly compare with siblings, but the context is clear. The restriction 'Python files only' provides a clear constraint, earning a 4.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Given function/class names in a .py file, return them verbatim (full source) plus their compressed dependency context.
| Name | Required | Description | Default |
|---|---|---|---|
| targets | Yes | Function/class names to expand. | |
| file_path | Yes | Path to a .py file (relative to --root or absolute). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations are absent, so description carries full burden. It states it returns 'full source' and 'compressed dependency context', which is a key behavioral trait. However, it does not disclose read-only nature, auth requirements, or potential performance implications (e.g., large files). Adequate but not comprehensive.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Single sentence, front-loaded with purpose. No wasted words; all information is necessary and directly conveys the tool's action and output.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given that there is no output schema and no annotations, the description is somewhat vague about the exact format of 'compressed dependency context'. While the tool is simple, the description could be more complete by clarifying what 'compressed' means. Sufficient but leaves room for ambiguity.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the schema already documents both parameters adequately. The tool description adds no new meaning beyond restating the schema descriptions, meeting the baseline but not adding value.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Description uses specific verb 'expand' on function/class names in a .py file, clearly indicating it returns full source plus dependency context. Differentiates from siblings like 'get_compressed' (which likely returns compressed content) and 'blast_radius' (impact analysis).
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No explicit guidance on when to use this tool versus alternatives like 'get_compressed' or 'blast_radius'. Usage is implied by the description (to get full source of specific definitions with dependencies), but lacks when-not-to-use or prerequisite information.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
S5-compress a single source file — Python or JS/TS (signatures + dependency contracts, bodies stripped) — and return the compressed structure plus token stats (tokens_before, tokens_after, reduction_pct).
| Name | Required | Description | Default |
|---|---|---|---|
| file_path | Yes | Path to a .py file (relative to --root or absolute). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations provided, so the description must disclose behaviors. It does not state whether the operation is read-only or modifies files, nor does it mention side effects, auth requirements, or rate limits. The word 'compress' might imply mutation, but the return of a 'compressed structure' suggests the original file is unchanged. This ambiguity is a gap.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single well-structured sentence. It front-loads the action ('S5-compress a single source file'), specifies file types, details the compression behavior, and lists the output. No unnecessary words.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Without output schema, the description should explain the return value structure, but it does mention 'compressed structure plus token stats (tokens_before, tokens_after, reduction_pct),' which is helpful. However, it lacks behavioral context (read/write), does not resolve the file type inconsistency, and does not address how this relates to sibling tools like 'expand.' The minimal viable information is present, but gaps reduce completeness.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% (one parameter with description). The description adds that the tool supports Python and JS/TS, but the schema's parameter description says 'Path to a .py file,' which contradicts the broader file types stated in the tool description. This inconsistency misleads the agent and detracts from the value added beyond the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool compresses a single source file (Python or JS/TS) by stripping bodies while keeping signatures and dependency contracts, and returns token stats. The verb 'compress' and resource 'source file' are specific, and it distinguishes from siblings like 'blast_radius' or 'expand' implicitly.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No explicit guidance on when to use this tool versus alternatives. The description does not mention when to choose compression over expansion or other sibling tools, leaving the agent without decision context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Walk a repo (Python + JavaScript/TypeScript) and return an aggregated structural index: every function, class, import and dependency. Max depth 3; skips pycache, .git, venv, node_modules.
| Name | Required | Description | Default |
|---|---|---|---|
| path | Yes | Repo path (relative to --root or absolute). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries full burden. It discloses maximum depth 3 and skipped directories (__pycache__, .git, venv, node_modules), but fails to mention behavior for unsupported file types, invalid paths, or performance implications for large repos. Additional context would improve transparency.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is two concise sentences with front-loaded purpose and clear constraints. Every word adds value, no redundancy.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given no output schema, the description partially explains return value as 'aggregated structural index' listing functions, classes, imports, dependencies. However, it lacks details on output format (e.g., JSON structure, list vs. dict) and does not address edge cases like empty repos or mixed languages. It is acceptable but not fully complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% with a single required 'path' parameter described adequately. The description adds no extra parameter details beyond the schema, so baseline 3 is appropriate. The description's main value is in explaining the tool's overall function, not parameter specifics.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool walks a repo and returns a structural index of functions, classes, imports, and dependencies, explicitly specifying supported languages (Python and JavaScript/TypeScript). This distinguishes it from sibling tools like 'blast_radius' or 'security_surface', which likely have different purposes.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description does not provide explicit guidance on when to use this tool versus alternatives like 'blast_radius' or 'expand'. It only implies usage for structural indexing, but lacks when-to-use, when-not-to-use, or prerequisites.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Walk a Python repo and find functions that call dangerous primitives — pure AST analysis, zero LLM calls. Classifies calls as HIGH risk (subprocess/os.system/eval/exec/pickle/import/ctypes …) or MEDIUM risk (sockets/requests/httpx/urllib/paramiko/smtplib and write-mode open()). Also scans for potential secrets: hardcoded keys/tokens (AKIA…, sk-…, ghp_…, token=/password=/api_key= assignments), .env/environment loading, and high-entropy token-looking literals. Returns high_risk, medium_risk, secrets, clean files and a summary. Max depth 3; skips pycache, .git, venv, node_modules.
| Name | Required | Description | Default |
|---|---|---|---|
| repo_path | Yes | Repo path (relative to --root or absolute). |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations provided, so description carries full burden. It clearly states pure AST analysis, zero LLM calls, risk classification levels, secret detection, max depth 3, and skipped directories. Could mention whether the tool is read-only (likely yes) but overall quite transparent.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Description is front-loaded with the main action and packed with relevant details in three sentences. Minimal redundancy, though length could be trimmed slightly without losing meaning.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given no output schema, description adequately explains return values (high_risk, medium_risk, secrets, clean files, summary). Covers key behaviors and limitations. Could mention file type filtering but sufficient for an agent to understand scope.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Only one parameter (repo_path) with 100% schema coverage. Description adds no extra detail beyond the schema's description of the path. Baseline score is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
Description clearly states it walks a Python repo using AST analysis to find dangerous function calls and secrets, with specific risk classifications. No sibling tool overlaps, making its unique purpose unmistakable.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies when to use (security scanning of Python repos) but does not explicitly state when not to use or mention alternatives among siblings. More precise guidance would improve this dimension.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
v0.7.0blast_radiusexpandget_compressedindex_reposecurity_surfaceScored across 5 tools
Each tool has a distinct purpose: impact analysis, source retrieval, file compression, repo indexing, and security scanning. No two tools overlap in functionality.
All tools use snake_case, but 'expand' is a lone verb while others follow a verb_noun pattern (e.g., get_compressed, index_repo). Minor inconsistency but still clear.
5 tools is a well-scoped set for a static analysis toolkit. Neither too few nor too many, each tool serves a clear need.
Covers core static analysis needs (impact, source, compression, indexing, security). Missing a dedicated diff or test analysis tool, but the surface is reasonably complete for the domain.
The OpenRouter for tools. One MCP connection gives any AI agent 254 hosted tools, pay per call.
One MCP endpoint for Claude, GPT & Gemini: 100+ tools + no-code connectors + agent workers.
Security & DLP proxy for MCP: tool-poisoning scans, PII redaction on tool args/results. Beta.
MCP server for your apps' tools and custom tools, plus hosted AI agents and approval-gated workflows
