sage

Sage AI智能体安全 — AI编码助手的代理检测与响应 Sage 是一个轻量级安全层，用于保护 AI 代理免于执行危险操作。它拦截工具调用（shell 命令、URL 获取、文件写入），并在执行前通过多层威胁检测进行检查。 注意： Sage 可能以不同的产品名称出现（例如 Norton Sage、Avast Sage），具体取决于安装方式。详见品牌。 主要特性 URL 信誉 — 基于云端的恶意软件、钓鱼和诈骗 URL 检测 本地启发式规则 — 300+ 个基于 YAML 的威胁模式，用于检测危险命令、可疑 URL、凭证泄露和混淆行为 提示注入检测 —...

MakerChecker

🛡️ MakerChecker AI 代理的开源安全层。 默认拒绝执行、人工审批以及加密签名的审计追踪——确保您的代理仅运行被授权的内容，并且可证明无法自行批准其自身的工作。 网站 · 在线演示 · mc scan · 文档 您的代理继续在其现有框架（LangChain、Claude SDK、CrewAI）中运行。MakerChecker 位于每个工具调用的前后：作为检查点在前，作为签名账本在后。代理仅通过角色 行动，仅执行被授予 的技能，不能超出其限制，也不能批准自身的工作。 🚀 快速入门 1 — 扫描您的代码...

nah

代价高昂的错误到此为止 一个阻止灾难性 agent 操作的守卫 nahguard.ai • 它阻止什么 • 它如何比较 • 它如何决策 • 安装 • 扩展 • 威胁模型 claude code · codex · cursor · pi · + 11 更多 nah 是一个守卫，位于你的编码 agent 的 hook 路径中，在工具调用运行之前读取它们。它会阻止那些它能证明是灾难的调用，并且从不批准任何东西：其他所有内容都交由你的运行时正常权限处理。 nah 只是一个 Rust 二进制文件：裁决是确定性的，不需要 LLM。 扩展只是程序。将你的 agent 指向 nah...

node9-proxy

🛡️ node9 Access control for AI agents Your AI agents can reach Slack, GitHub, email, and your database. node9 decides what they may do with each one. Credential jail · secrets and PII · destructive git, SQL and shell held for review · MCP tool pinning · network egress allowlist · loop breaker · on...

scopeblind-gateway

protect-mcp Fail-closed Cedar policy gate plus signed receipts for AI agent tool calls. protect-mcp is a gate that sits in front of an AI agent's tool calls. It evaluates each call against a Cedar policy, blocks what breaks the rules before it runs, and signs an offline-verifiable Ed25519 receipt...

ROPE

ROPE: Aplicación de Política de Origen Enrutado Código fuente de nuestro artículo: ROPE: Aplicación de Política de Origen Enrutado contra la Inyección Indirecta de Prompts por Xinhang Ma, Chaowei Xiao, William Yeoh, Ning Zhang, Yevgeniy Vorobeychik Resumen La inyección indirecta de prompts IPI...

PYSEC-2026-4124 Open WebUI: A user's session cookies are sent to tool servers configured for bearer authentication

Summary When more than one external tool server is reachable in the same request, a tool call to a server configured for bearer authentication can arrive carrying the calling user's Open WebUI session cookies alongside that server's own key. The cookie jar is built per connection, but the callabl...

PT-2026-105873

Summary When more than one external tool server is reachable in the same request, a tool call to a server configured for bearer authentication can arrive carrying the calling user's Open WebUI session cookies alongside that server's own key. The cookie jar is built per connection, but the callabl...

sage v0.13.0

Sage Safety for Agents — Agent Detection & Response for AI coding assistants Sage is a lightweight security layer that protects AI agents from executing dangerous actions. It intercepts tool calls — shell commands, URL fetches, file writes — and checks them against multiple threat detection layer...

PYSEC-2026-3946 Xinference vulnerable to remote code execution via unsafe `eval()` in Llama3 tool-call parsing

Summary Xinference used Python's unsafe eval function when parsing Llama3 tool-call output generated by a large language model. Because the model output can be influenced by attacker-controlled prompts sent to the chat completion API, a remote attacker can craft prompts that cause the model to...

CVE-2026-78230

AshAi exposes Ash read actions to language-model tool calls. The read tool accepts an aggregate result type min, max, sum, avg that builds an ad-hoc Ash.Query.Aggregate over a named field and returns its raw value. Ash field policies redact forbidden fields on returned records replacing them with...

CVE-2026-78230 AshAi aggregate tool can read field-policy-protected fields

AshAi exposes Ash read actions to language-model tool calls. The read tool accepts an aggregate result type min, max, sum, avg that builds an ad-hoc Ash.Query.Aggregate over a named field and returns its raw value. Ash field policies redact forbidden fields on returned records replacing them with...

PT-2026-87175

AshAi exposes Ash read actions to language-model tool calls. The read tool accepts an aggregate result type min, max, sum, avg that builds an ad-hoc Ash.Query.Aggregate over a named field and returns its raw value. Ash field policies redact forbidden fields on returned records replacing them with...

Improper Authorization

Spring AI is vulnerable to Improper Authorization. The vulnerability is due to per-request tool restrictions not being fully enforced during tool-call dispatch, allowing a tool unavailable to the current request to be invoked and potentially enabling privilege escalation...

CVE-2026-82579 AshAi tool loop never terminates when all tool calls are filtered out, enabling denial of service

Loop with Unreachable Exit Condition Infinite Loop vulnerability in ash-project ashai allows an attacker who can influence a model's output to hang the tool loop and drive unbounded, repeated model requests. AshAi.ToolLoop classifies a model response of :toolcalls, then filters the calls through...

PT-2026-83668

Loop with Unreachable Exit Condition Infinite Loop vulnerability in ash-project ash ai allows an attacker who can influence a model's output to hang the tool loop and drive unbounded, repeated model requests. AshAi.ToolLoop classifies a model response of :tool calls, then filters the calls throug...

mcpsnoop v0.19.0

Wireshark for MCP. A transparent proxy that shows every real tool call between your AI client and your MCP servers, live in your terminal. The problem The official MCP Inspector connects as its own client, so it never sees what your client Cursor, Claude Code, Codex actually sends your server. An...

Xinference vulnerable to remote code execution via unsafe `eval()` in Llama3 tool-call parsing

Summary Xinference used Python's unsafe eval function when parsing Llama3 tool-call output generated by a large language model. Because the model output can be influenced by attacker-controlled prompts sent to the chat completion API, a remote attacker can craft prompts that cause the model to...

CVE-2026-61539 Xinference: Remote code execution via unsafe `eval()` in Llama3 tool-call parsing

Xinference is an inference API for running open-source, speech, and multimodal models. In 2.5.0 and earlier, Xinference passes attacker-influenced Llama3 tool-call output to eval in xinference/model/llm/toolparsers/llama3toolparser.py and xinference/model/llm/utils.py. Requests to...

CVE-2026-61539

Xinference is an inference API for running open-source, speech, and multimodal models. In 2.5.0 and earlier, Xinference passes attacker-influenced Llama3 tool-call output to eval in xinference/model/llm/toolparsers/llama3toolparser.py and xinference/model/llm/utils.py. Requests to...