sdk

@cosignet/sdk Thin, zero-dependency TypeScript client for Cosignet — human-in-the-loop approval for high-risk AI-agent actions, with payload-bound passkey signatures. Put a human in the loop before a dangerous action runs: pause it, get an explicit passkey approval from a person Face ID / Touch I...

npm: obsidian-tc: Declined human-in-the-loop confirmations can permit destructive calls over HTTP

npm: obsidian-tc: Declined human-in-the-loop confirmations can permit destructive calls over HTTP vulnerability discovered by ? in WordPress npm obsidian-tc versions = 1.13.0, = 1.31.3...

EUVD-2026-109082

obsidian-tc is a governed, agent-ready Model Context Protocol server for Obsidian vaults. From 1.13.0 until 1.31.4, on a Streamable HTTP deployment with auth.jwtSecret configured and a 2026-07-28-era client declaring elicitation.form support, a dispatch-gated tool call offered an inputRequired...

CVE-2026-115477 obsidian-tc: Declined human-in-the-loop confirmations can permit destructive calls over HTTP

obsidian-tc is a governed, agent-ready Model Context Protocol server for Obsidian vaults. From 1.13.0 until 1.31.4, on a Streamable HTTP deployment with auth.jwtSecret configured and a 2026-07-28-era client declaring elicitation.form support, a dispatch-gated tool call offered an inputRequired...

CVE-2026-115477

The obsidian-tc Model Context Protocol server (versions 1.13.0 through 1.31.4 ) contains a vulnerability where destructive tool calls can be executed without human approval. In Streamable HTTP deployments with auth.jwtSecret configured and specific client support, the server improperly trusted a ...

TamarinAgent

Confidence-Guided Protocol IR Review UI This repository contains a local human-in-the-loop UI for LLM-aided security protocol modeling. It supports the workflow described in our work: Confidence-Guided Protocol IR for LLM-Aided Security Protocol Modeling The system introduces a protocol...

GitGot

Description GitGot is a semi-automated, feedback-driven tool to empower users to rapidly search through troves of public data on GitHub for sensitive secrets. How it Works During search sessions, users will provide feedback to GitGot about search results to ignore, and GitGot prunes the set of...

PT-2026-122122

obsidian-tc is a governed, agent-ready Model Context Protocol server for Obsidian vaults. From 1.13.0 until 1.31.4, on a Streamable HTTP deployment with auth.jwtSecret configured and a 2026-07-28-era client declaring elicitation.form support, a dispatch-gated tool call offered an inputRequired...

Confidence-Guided Protocol IR for LLM-Aided Security Protocol Modeling

Large language models offer a promising interface for translating natural-language protocol descriptions into formal security models, but their outputs remain difficult to trust without expert validation. In this paper, we present a human-in-the-loop framework for generating Tamarin-verifiable...

PYSEC-2026-3707

Apache Airflow 3.3.0 moved human-in-the-loop tasks from the triggerer to a new awaitinginput task state swept by the scheduler. That sweep deserializes the task instance's nextkwargs without an allow-list, so a Dag author — who controls that value through the task execution API — can cause an...

EUVD-2026-57275

Apache Airflow 3.3.0 moved human-in-the-loop tasks from the triggerer to a new awaitinginput task state swept by the scheduler. That sweep deserializes the task instance's nextkwargs without an allow-list, so a Dag author — who controls that value through the task execution API — can cause an...

CVE-2026-67260

Apache Airflow 3.3.0 moved human-in-the-loop tasks from the triggerer to a new awaitinginput task state swept by the scheduler. That sweep deserializes the task instance's nextkwargs without an allow-list, so a Dag author — who controls that value through the task execution API — can cause an...

PraisonAI: HTTPApproval dashboard renders tool arguments as raw HTML, allowing approval-page XSS to approve dangerous tools

HTTPApproval dashboard renders tool arguments as raw HTML, allowing approval-page XSS to approve dangerous tools Summarypraisonai.bots.HTTPApproval renders pending tool approval arguments directlyinto the approval dashboard HTML. An attacker-controlled tool argument caninject JavaScript into that...

GARAGE: Characterizing the Automation Boundary in LLM-Based Attack Graph Generation

While modern vehicle security depends on effective Cyber Threat Intelligence CTI synthesis, current automated tools struggle with unstructured data and automotive-specific architectural nuances. To bridge this gap, we introduce GARAGE, a RAG-powered framework that converts fragmented CTI into an...

PraisonAI: Unauthenticated Allow-List Manipulation Bypasses Agent Tool Approval Safety Controls

SummaryThe gateway's /api/approval/allow-list endpoint permits unauthenticated modification of the tool approval allowlist when no authtoken is configured the default. By adding dangerous tool names e.g., shellexec, filewrite to the allowlist, an attacker can cause the ExecApprovalManager to...

PT-2026-59538

Summary The gateway's /api/approval/allow-list endpoint permits unauthenticated modification of the tool approval allowlist when no auth token is configured the default. By adding dangerous tool names e.g., shell exec, file write to the allowlist, an attacker can cause the ExecApprovalManager to...

CVE-2026-30911

Apache Airflow versions 3.1.0 through 3.1.7 missing authorization vulnerability in the Execution API's Human-in-the-Loop HITL endpoints that allows any authenticated task instance to read, approve, or reject HITL workflows belonging to any other task instance. Users are recommended to upgrade to...

GHSA-8X34-9Q3V-H7G8 Apache Airflow: Execution API HITL Endpoints Missing Per-Task Authorization

Apache Airflow versions 3.1.0 through 3.1.7 missing authorization vulnerability in the Execution API's Human-in-the-Loop HITL endpoints that allows any authenticated task instance to read, approve, or reject HITL workflows belonging to any other task instance. Users are recommended to upgrade to...

PYSEC-2026-17

Apache Airflow versions 3.1.0 through 3.1.7 missing authorization vulnerability in the Execution API's Human-in-the-Loop HITL endpoints that allows any authenticated task instance to read, approve, or reject HITL workflows belonging to any other task instance. Users are recommended to upgrade to...

CVE-2026-30911

Apache Airflow versions 3.1.0 through 3.1.7 missing authorization vulnerability in the Execution API's Human-in-the-Loop HITL endpoints that allows any authenticated task instance to read, approve, or reject HITL workflows belonging to any other task instance. Users are recommended to upgrade to...