augustus v0.14.30

Augustus - LLM vulnerability scanner for prompt injection, jailbreak, and adversarial attack testing Augustus - LLM Vulnerability Scanner Test large language models against 210+ adversarial attacks covering prompt injection, jailbreaks, encoding exploits, and data extraction. Augustus is a Go-bas...

Red-Teaming Auto Mode: Improving Blocking Classifiers against Malign Coding Agents

To keep coding agents from going off the rails, production systems now review each proposed action with a blocking monitor that can reject it before it runs Auto Mode in Claude Code, Guardian in OpenAI's Codex. Prior evaluations of such monitors largely measure robustness to accidental harm or...

Measuring and Exploiting Implicit Trust in LLM Tool-Calling Pipelines

The Model Context Protocol MCP enables LLMs to invoke external tools, but every tool interaction exposes the model to attacker-controlled text through multiple input channels tool descriptions, tool results, sampling messages that share a single context window without privilege separation. In thi...

reasongate v0.4.0

ReasonGate A self-hostable gate that inspects the text going into and out of an LLM and returns an explainable allow / flag / block decision with a machine-readable audit record for every call. What this is The open-source core is rule-based. It does four things: recognizes known prompt-injection...

GHSA-5648-RGJ9-V224 @zereight/mcp-gitlab has multiple safety-control bypasses: execute_graphql read-only + allow-list bypass, unauthenticated transports, session-exhaustion DoS

Summary @zereight/mcp-gitlab exposes GitLab to an LLM agent while relying on read-only mode, a project allow-list, and transport auth as its safety controls. Five defects defeat those controls. Under the MCP threat model, tool-call arguments/content can be shaped by untrusted input prompt injecti...

CVE-2026-53957 Contentful MCP Server: export_space/import_space tools pass LLM-controlled `host`/`proxy` args to CMA client, redirecting server PAT to attacker-controlled endpoint

Contentful MCP Server is a Model Context Protocol server for the Contentful Management API. Prior to @contentful/mcp-server 1.7.19 and @contentful/mcp-tools 0.4.5, exportspace and importspace in packages/mcp-tools/src/tools/jobs/space-to-space-migration/exportSpace.ts and...

CVE-2026-53957 Contentful MCP Server: export_space/import_space tools pass LLM-controlled `host`/`proxy` args to CMA client, redirecting server PAT to attacker-controlled endpoint

Contentful MCP Server is a Model Context Protocol server for the Contentful Management API. Prior to @contentful/mcp-server 1.7.19 and @contentful/mcp-tools 0.4.5, exportspace and importspace in packages/mcp-tools/src/tools/jobs/space-to-space-migration/exportSpace.ts and...

CVE-2026-53957

CVE-2026-53957 affects the Contentful MCP Server (@contentful/mcp-tools and @contentful/mcp-server). The export_space and import_space tools spread LLM-controlled host, proxy, rawProxy, and insecure arguments into the options passed to the Contentful Management API SDK, which then attaches the se...

giskard-oss giskard-checks/v1.0.4

Evals, Red Teaming and Test Generation for Agentic Systems Modular, Lightweight, Dynamic and Async-first Docs • Website • Community !IMPORTANT Giskard v3 is a fresh rewrite designed for dynamic, multi-turn testing of AI agents. This release drops heavy dependencies for better efficiency while...

CVE-2026-53957: Server-Side Request Forgery (SSRF)

Contentful MCP Server is a Model Context Protocol server for the Contentful Management API. Prior to @contentful/mcp-server 1.7.19 and @contentful/mcp-tools 0.4.5, exportspace and importspace in packages/mcp-tools/src/tools/jobs/space-to-space-migration/exportSpace.ts and...

CVE-2026-84601

A permissions issue was addressed with improved state management. This issue is fixed in macOS Golden Gate 27. An app may be able to bypass Apple Intelligence security prompts...

CVE-2026-84601

A permissions issue was addressed with improved state management. This issue is fixed in macOS Golden Gate 27. An app may be able to bypass Apple Intelligence security prompts...

CVE-2026-84601

A permissions issue was addressed with improved state management. This issue is fixed in macOS Golden Gate 27. An app may be able to bypass Apple Intelligence security prompts...

EUVD-2026-77913

A permissions issue was addressed with improved state management. This issue is fixed in macOS Golden Gate 27. An app may be able to bypass Apple Intelligence security prompts...

memory-poisoning-axis

memory-poisoning-axis Deterministic memory-poisoning / prompt-injection measurement — anchored to CVE-2026-24301 "CoSnitch" Microsoft Copilot Personal, CVSS 3.1 8.8, patched 18 Aug 2026. CoSnitch chained: i automatic prompt execution ?q= + undocumented ?autorun=1, ii data exfiltration via...

CVE-2026-57145 PraisonAI: Arbitrary File Read/Write via `multiedit` Tool Without Path Validation

PraisonAI is a multi-agent teams system. Prior to 4.6.62, src/praisonai/praisonai/tools/multiedit.py passes the LLM-controlled filepath parameter directly to open for reading and writing without traversal rejection, symlink resolution, a workspace boundary, or protected-path checks...

CVE-2026-57145 PraisonAI: Arbitrary File Read/Write via `multiedit` Tool Without Path Validation

PraisonAI is a multi-agent teams system. Prior to 4.6.62, src/praisonai/praisonai/tools/multiedit.py passes the LLM-controlled filepath parameter directly to open for reading and writing without traversal rejection, symlink resolution, a workspace boundary, or protected-path checks...

EUVD-2026-77614

PraisonAI is a multi-agent teams system. Prior to 4.6.59, the CODETOOLS wrappers keep workspaceroot as None and pass workspace=None to readfile, searchreplace, and applydiff helpers that enforce path containment only for a truthy workspace. An application that exposes codereadfile,...

EUVD-2026-77543

PraisonAI is a multi-agent teams system. Prior to praisonaiagents 1.6.59, MentionsParser.processfilemention accepts file-mention values and falls back from workspace-relative resolution to Pathfilepath without traversal, symlink, or workspace-boundary validation. Prompt input from users, bots, or...

EUVD-2026-77541

PraisonAI is a multi-agent teams system. Prior to praisonaiagents 1.6.59, executecode sandbox mode permits runtime assembly of blocklisted dunder names and allows str.format or str.formatmap to resolve dotted fields through C-level attribute access that bypasses safegetattr. This exposes class,...