Skip to main content
Search

# Model Context Protocol Server

6 related articles

Subscribe to updates
  1. Claude Security rescans as code changes and closes findings when fixes ship, replacing scheduled scans

    Claude Security has moved from on-demand scanning to continuous coverage: after a repository is connected it runs a deep baseline scan, then rescans changes according to their risk and complexity, and marks findings fixed when the fix lands. Previously a scan captured one moment of the code, leaving teams to schedule scans themselves. Each project also carries a knowledge base and an editable threat model that shape what gets flagged.

  2. Claude Platform opens OpenTelemetry export: trace Messages API calls and managed-agent sessions in your own backend

    Claude Platform now exports tracing data for Messages API requests and Claude Managed Agents sessions as OpenTelemetry spans, so developers can view those calls in their own observability backend.

  3. NVIDIA Blog

    RTX Spark laptops ship Friday: a 74B agent model squeezed into 24GB runs locally

    NVIDIA says RTX Spark laptops arrive Oct. 16 with a Nemotron 3.5 Bolt 74B model quantized to 2 bits and packed into 24GB, letting local agents run on a single machine.

  4. arXiv

    APEX moves indirect prompt-injection defense to the execution boundary, cutting attack success to 0% on five of six benchmarks and 0.56% on the sixth

    The work presents APEX, which shifts indirect prompt-injection defense from recognizing attack patterns to checking each proposed effect at the execution boundary: before untrusted execution it compiles the user's authorization into a contract, WRAP admits an effect only when the contract and runtime evidence reconstruct its authorization, and PLANT places probes on the contract's endorsed dependencies so unendorsed use of runtime information reveals itself before commitment; across six benchmarks and 13 baselines APEX reaches 0% attack success on five benchmarks and 0.56% on the sixth, and holds 0% under three adaptive attacks spanning all three capability-unit types.

  5. arXiv

    CyTReX turns DNP3 anomaly alerts into traceable ranked threat hypotheses, with a five-configuration comparison showing the complete pipeline gives the richest evidence but weaker label anchoring

    The paper presents CyTReX, an explainable threat reasoning framework for distributed energy resource (DER) networks that consolidates edge anomaly detection evidence, cloud attack interpretation, SHAP feature attributions, surrogate decision rules, and MCP-enabled cyber threat intelligence into a unified evidence packet that constrains large language model reasoning to produce ranked threat hypotheses and attack trees; across five configurations on a DNP3 case study, adding reasoning components improved hypothesis specificity, evidence traceability, and analytical grounding, with the complete pipeline providing the richest evidence-grounded reasoning context.

  6. LLM multi-agent control framework reaches a 93% mean solve rate in a six-module simulated factory, with the orchestrator autonomously rerouting around a silent conveyor-belt fault in all ten runs

    The work pairs each factory module with a dedicated LLM-based agent and an MCP tool server that exposes the module's skills via OPC UA method calls, with agents coordinating over MQTT and grounded by real-time factory-state updates; in a simulation of a six-module hexagonal factory it compares orchestrator, peer-to-peer, and monolithic architectures across nine production challenges of increasing complexity, finding that monolithic and peer-to-peer both achieve the highest mean solve rate (93%) while the orchestrator uniquely resolves a silent conveyor-belt fault in all ten runs by autonomously rerouting plates around the blocked segment, and that all architectures exhibit emergent fault-diagnosis behavior without explicit failure-handling logic.

You’re all caught up