← All posts

Platform & Architecture

338 posts on platform & architecture.

C2PA Content Credentials: An Implementation and Handoff Checklist

C2PA Content Credentials bind signed provenance claims to media through manifests and assertions. This implementation checklist covers assertion design, manifest creation, signing identity, trust configuration, validation, export handoff, and operating ownership. It also defines the boundary between media provenance tooling and an HTTP AI policy gateway.

ai-governanceai-securityarchitectureauditpolicy-enforcement
Read post →

Azure AI Foundry Security: What Entra ID, Private Endpoints, and RBAC Cover

Microsoft Entra ID, private endpoints, and RBAC secure three separate layers of an Azure AI Foundry deployment: authentication, network isolation, and resource-management permissions. None of the three evaluates whether a specific authenticated end user should be allowed to send a specific prompt to a specific model right now.

ai-securitycloud-securityllm-securityidentity-and-authorizationarchitecturezero-trust
Read post →

LLM JSON Schema Validation: Structure, Policy, and Scope

LLM JSON schema validation confirms structure, required fields, and types. This guide shows the separate policy checks for business rules, personal data, authorization scope, and tool-call arguments on an LLM response.

llm-engineeringjson-schemaai-engineeringstructured-outputai-security
Read post →

On-Prem LLM Security: A Deployment Acceptance Matrix

On-prem LLM security spans the GPU host and Kubernetes cluster. It also covers the inference server, model assets, and every exposed request route. A defensible launch decision assigns each surface to an owner, then pairs it with a test and retained evidence. This guide provides an acceptance matrix and keeps DeepInspect limited to authenticated HTTP model traffic deliberately routed through its policy point.

ai-securityllm-securityself-hosted-aidevsecopsauditarchitecture
Read post →

Microsoft PyRIT Red Team Setup: An Implementation Lab

Microsoft PyRIT is an open-source framework for repeatable generative AI red teaming across configurable targets and attack strategies, with scorers and memory. A useful setup lab proves one authorized target, records each conversation, exports the evidence, and replays a repaired case. This guide covers installation choice, target configuration, scorer and memory decisions, route scoping, evidence export, and the boundary between a test runner and a production policy layer.

ai-securityllm-securityprompt-injectiondevsecopsauditarchitecture
Read post →

Azure AI Red Teaming Agent: Deployment and Acceptance Runbook

The Azure AI Red Teaming Agent in Microsoft Foundry runs PyRIT-based adversarial scans and evaluates attack responses with Foundry risk and safety evaluators. Teams can run predeployment exercises or schedule postdeployment work, then review attack success rates and scorecards. This runbook covers prerequisites, role assignment, target and region checks, taxonomy review, evidence, remediation replay, and the boundary between a scanner and production enforcement.

ai-securityllm-securityprompt-injectiondevsecopsauditarchitecture
Read post →

Anthropic MCP Server Security: The Enterprise Controls That Sit Around Claude's Tool Layer

Anthropic's Model Context Protocol implementation lets Claude call tool servers with a standard schema. The enterprise question is what security controls have to sit around the MCP server when the LLM behind the protocol is Claude. This piece walks through the controls: transport authentication to the MCP server, token exchange between Claude's session and the tool call, response inspection before the tool result reaches the model, and audit records tied to the human user who authorized the session.

anthropicmcpclaudeai-securityai-agentai-gateway
Read post →

MCP Server Authorization Patterns: Enforcing Who Can Call Which Tool Through the Model Context Protocol

The Model Context Protocol gave agents a common way to talk to tool servers. It did not give the enterprise a common way to authorize which agent can call which tool with which arguments. Authorization at the MCP server sits at three layers: the transport layer, the server layer, and the tool-invocation layer. Each layer answers a different question and produces a different audit record. This piece walks through the authorization patterns that survive an enterprise deployment across multiple MCP servers and multiple agent identities.

mcpmodel-context-protocolauthorizationai-agentai-securityai-gateway
Read post →

MCP Server Authentication: Identity Binding at the Model Context Protocol Boundary

The Model Context Protocol lets an LLM client discover and call tools exposed by an MCP server. Authentication at the MCP boundary determines which identity issues the tool calls, which policy applies, and which record ends up in the audit log. This piece walks through the OAuth 2.1 authorization flow the MCP spec adopted, the pitfalls in shared-secret patterns, and the inspection-layer architecture that binds every MCP tool call to a verified identity.

mcpauthenticationoauthai-agent-securityai-engineering
Read post →

Replit Agent Security: Four Control Boundaries to Test

Replit assigns Agent harness security to its platform while customers own generated-code review, prompt hygiene, sensitive-action approval, and third-party Skill or MCP vetting. Agent security scans add source and static analysis, but Replit calls them incomplete. A credible review tests the Agent, generated application, deployment, and routed model-request boundaries separately.

ai-securityagentic-aidevsecopszero-trustpolicy-enforcement
Read post →

AWS Bedrock Security: Guardrails, IAM, and Privacy Gaps

AWS Bedrock security uses Guardrails for content, PrivateLink for a private API path, and IAM for API access. This guide shows what each control covers, what Bedrock privacy settings govern, and where per-request authorization remains with the application or a separate policy layer.

ai-securitycloud-securityllm-securityidentity-and-authorizationarchitecturezero-trust
Read post →