Protect Mcp

io.github.tomjwxf由 GitHub 账号 tomjwxf 发布v0.5.3更新于 Oct 24, 2026

Signed receipts and Cedar policies for AI agent tool calls. Claude Code hooks, MCP gateway.

已验证STDIO仅桌面开发者工具安全与监控

概览

AI 生成的概览

一个本地策略网关,按 Cedar 规则检查 AI 智能体的工具调用,拦截违规调用,并为每次决策签发可验证的签名回执。

可以做什么
按 Cedar 策略评估工具调用并放行或拒绝,出错时默认拒绝,执行模式启动前先做自检。包装 MCP 服务器或 Claude Code 钩子,让每次调用都经过网关,并可根据观察到的调用生成可审阅的策略草稿。将每次决策签名为 Ed25519 回执,在本地查看、导出或离线验证回执与审计包。在网关中保管凭据,智能体只看到标签,实际值在派发时注入请求头、查询参数或环境变量。可选地将回执摘要锚定到托管服务,并使用协作工具处理共享发票或代码库评审任务。
适用场景
当你运行可调用工具的 AI 智能体,希望有一道本地、出错即拒绝的防线来拦截违规操作并留下签名证据时使用。也适合需要可离线验证的审计记录,或希望让 API 密钥不进入智能体上下文的场景。
运行要求
以 npm 包形式通过 npx 在本地运行(需要 Node.js)。初始化会生成 Ed25519 密钥对和策略文件。可选的托管功能需要 ScopeBlind 账户、SCOPEBLIND_TOKEN 环境变量和网络访问;网关自身从进程环境变量读取密钥。
安装前请注意
网关会拦截、拒绝或放行工具调用,并向外部请求注入凭据,从影子模式切换到执行模式前应审阅策略。0.5.x 和 0.6.x 版本出错时放行且 Cedar 评估不正确,应使用 0.7.0 或更高版本。协作功能会把草稿文本和私有说明发送给 ScopeBlind,托管锚定会上传回执摘要和计费元数据。私有配置文件保存智能体密钥,必须妥善保密。

安装

在 SourceWeft 中

  1. 打开 控制台中的 Protect Mcp,将其添加到工作区。
  2. 为需要使用其工具的对话启用该服务。

Desktop only,通过 STDIO。 STDIO 服务会启动本地进程,因此需要 SourceWeft 桌面宿主。

其他 MCP 客户端

参照 仓库 中的启动说明。

README

protect-mcp

Local policy enforcement and signed decision receipts for AI agent tool calls.

[npm version] [Monthly npm downloads] [Node.js compatibility] [Package checks] [MIT license]

Developer page · Quickstart · Agent setup · Policy packs · Sample record · Integrations · Documentation · Contribute

Put the gateway between an MCP client and its tool server, or use its hooks in a coding agent. Review a Cedar policy, enable enforcement, and keep a signed record of the decisions. Local evaluation and verification need no ScopeBlind account and send no decision telemetry.

  • Enforce at the tool boundary. Cedar denies on missing policies, engine errors or evaluation errors in fail-closed mode. Shadow mode is the default for the MCP wrapper: use --enforce when ready to block.
  • Check what was signed. Ed25519 receipts include the decision and policy digest. @veritasacta/verify checks the bytes against the public key you trust.
  • Start with existing tools. Use MCP stdio, the HTTP gateway or coding-agent hooks. Optional hosted collaboration has its own permissions and data flows.

A gateway covers calls routed through it. It is not an OS sandbox, and a signed receipt does not establish that an action executed or that no action was omitted. The security policy describes the boundary and reporting process.

See the result

A real local run: evaluate two proposed calls, sign the decisions, verify the receipts, then change a signed decision and check it again.

[Five-step result summary: a read is allowed, a proposed force-push is denied, both decisions are signed, their receipts verify, and a modified denial fails signature verification. No proposed tool is executed.]

Watch the 14-second walkthrough

[Animated result summary of the same five steps, based on captured CLI output.]

Text walkthrough and reproduction · Captured results

Try it without an agent

Use Node.js 22 and npm in a new directory. Install the pinned gateway and verifier:

sh
mkdir protect-mcp-democd protect-mcp-demonpm init -ynpm install --save-exact [email protected] @veritasacta/[email protected]

This example evaluates two proposed calls and signs their decisions. It does not execute either tool or run git push. Everything after installation is local.

sh
# Create a starter policy and a private signing key.npx --no-install protect-mcp init --starter
# A normal read is allowed; a force-push to main is denied (exit 2).npx --no-install protect-mcp evaluate --policy ./protect.cedar --tool Read --input '{"path":"README.md"}'npx --no-install protect-mcp evaluate --policy ./protect.cedar --tool Bash --input '{"command":"git push --force origin main"}' || test "$?" -eq 2
# Evaluate again at signing time and record both decisions, in order.npx --no-install protect-mcp sign --cedar builtin:starter --tool Read --input '{"path":"README.md"}' --key ./protect-mcp.key --receipts ./receiptsnpx --no-install protect-mcp sign --cedar builtin:starter --tool Bash --input '{"command":"git push --force origin main"}' --key ./protect-mcp.key --receipts ./receipts
# Take only the public key from the key you generated, not from the receipt.PROTECT_MCP_PUBLIC_KEY=$(node -p 'JSON.parse(require("fs").readFileSync("protect-mcp.key", "utf8")).publicKey')npx --no-install verify --replay-chain ./receipts/receipts.jsonl --key "$PROTECT_MCP_PUBLIC_KEY" --json
# Change the signed deny to allow. The original receipt log stays intact.node -e 'const fs=require("fs");const r=JSON.parse(fs.readFileSync("receipts/receipts.jsonl","utf8").trim().split("\n")[1]);r.payload.decision="allow";fs.writeFileSync("tampered.json",JSON.stringify(r))'npx --no-install verify tampered.json --key "$PROTECT_MCP_PUBLIC_KEY" --json || test "$?" -eq 1

Expected: allow, then deny; both original receipt signatures and their link verify; the modified receipt reports invalid_signature. Keep protect-mcp.key private. To verify someone else's record, obtain their expected public key through an independent trusted channel.

Verifier exit codes are 0: valid, 1: invalid, 2: not checked/error. An invalid result is not necessarily tampering: the wrong expected key can also fail verification. Chain replay checks the supplied links; it does not establish that you received the entire history.

CI runs this exact example against the packed candidate and the pinned published verifier. See validation and releases.

Connect your agent

Choose the integration that matches your host and the coverage you need. Installing guidance skills does not enable enforcement. doctor checks local configuration; confirm the selected route with a real allowed and blocked call.

For the Claude Code plugin, first prepare and start the local gate, then install it in your project (Claude Code 2.1.295 or later):

sh
claude plugin install protect-mcp --marketplace ScopeBlind/scopeblind-gateway --scope project

The plugin blocks routed tool calls while its gate is unavailable. Choose the plugin or standalone project hooks, so the same call is not recorded twice.

Use caseStart here
Wrap an existing MCP serverMCP setup and shadow-to-enforce workflow
Claude Code enforcement pluginPlugin setup and local gate
Claude Code project hooksHook setup
Cursor project previewExplicit setup, tool mapping and host checks
ChatGPT / Codex assistant previewSetup guidance and a harmless verification demo, with no hooks
Existing Claude directory guidance pluginPublished plugin source, separate from enforcement
Call the policy evaluator as MCP toolsThe gate as an MCP server
Host an HTTP gatewayHTTP deployment notes

Host compatibility and release status separates adapter tests from real-host validation. The Cursor and assistant packages are previews, not public directory listings. The standalone hook-format fixes ship in 0.32.4; its changelog lists what changed for each format. A format flag alone does not establish a working Codex, Gemini or Hermes integration.

For an existing MCP server, initialize its configuration in a separate directory from the example above, then run the wrapper with your server command:

sh
npx --yes [email protected] initnpx --yes [email protected] --policy ./protect-mcp.json -- node your-mcp-server.js

This starts in shadow mode. Review the observed calls and policy before adding --enforce. wrap prints client configuration and exits; it is not the launch command. Full setup instructions.

Starter shell rules inspect command text. They are useful guardrails, but can miss alternate command spellings and can match harmless quoted text. Review the starter policy's coverage and limits before relying on it.

Integrations and contributions

ProjectConnection
Microsoft AI Agents for BeginnersEducational contribution. protect-mcp's maintainer authored Lesson 18 on signing, verifying and chaining agent receipts.
Cedar for AgentsUpstream contribution. Contributed the WASM bindings that expose Cedar's Rust MCP schema generator to JavaScript and TypeScript.
wshobson/agentsDownstream integration. Ships a protect-mcp plugin with Claude Code policy and receipt hooks.
TRACEInteroperability checks. Publishes Acta decision-receipt fixtures; our verifier reports its results against a pinned snapshot.

Sources, versions and check coverage.

Upgrading from 0.31

0.32.0 closes several ways around the gate. Some of them change how you start it:

  • --http listens on 127.0.0.1, and every request except GET /health needs Authorization: Bearer <token>. Set the same PROTECT_MCP_HTTP_TOKEN (at least 32 characters) for the gateway and for the client or ingress that calls it. Use --host only when you mean to expose the port beyond this machine.
  • --enforce refuses to start when signing is configured but its key is missing or unreadable, and refuses an allowed call whose receipt cannot be signed.
  • --cedar with --policy applies both: a call needs a Cedar permit and must also pass that file's block, require_approval, rate limits and min_tier. Signing and credentials come from that file.
  • The dashboard accepts a change only from its own page, which carries a per-launch token.
  • Signing keys are written 0600 in a 0700 folder. For a key made by an earlier version, run chmod 600 keys/gateway.json; the gateway warns until you do.
  • A once approval at POST /approve names the exact action: the Request ID the gateway prints, tool:payload_hash. The approval nonce is on the gateway's stderr only.

See the changelog for every change.

Documentation

Contribute

Start with a bounded help-wanted task: one real host capture, another clean-package runtime, or one tested policy recipe.

Bug reports, minimal reproductions, documentation fixes and compatibility tests are welcome. See CONTRIBUTING.md for setup and checks, and GOVERNANCE.md for maintainership. Report vulnerabilities privately through SECURITY.md, not a public issue.

The shared conformance corpus records compatibility results across receipt producers. A passing corpus is bounded interoperability evidence, not a security certification.

MIT licensed; see LICENSE and the existing patent notice. Maintained by Tom Farley at ScopeBlind. Receipt format: Acta individual Internet-Draft.

来源:README.md,提交 0736060

工具

0
SourceWeft 尚未在沙箱中启动过该软件包,因此其工具未知。

版本历史

1
  1. v0.5.3最新Sep 16, 2026

评分

已满足适用检查项的 50%

该评分衡量此收录条目的完整程度,依据注册中心条目、端点与公开来源计算得出。它不是对服务器本身的评价,也不表示服务器安全或当前可用:可连通性是带日期的检查结果,评分也不包含安装量或用户评价。

  • 近 30 天内可连通25 分

    请保持端点能够响应 MCP 握手;SourceWeft 每 30 天探测一次。

  • 已声明认证方式10 分

    请在注册中心条目中声明凭据请求头,或保持端点能够响应探测,以便判定其认证方式。

  • 已列出工具15 分

    请让服务器在无凭据时响应 tools/list,或发布可在沙箱中无凭据启动的软件包。

  • 描述必需15 分
  • 图标10 分
  • 文档15 分
  • 许可证10 分