AgntID Blog

Technical GuideSeptember 30, 2026

AWS Bedrock Agent Security: How to Secure Action Groups, IAM Roles, and Knowledge Bases

AWS Bedrock agents can reason over user requests, query knowledge bases, and invoke business actions through action groups. Production security starts with what the agent can call, what permissions back those actions, and what data the agent can retrieve.

SH

Sachin HHead of Marketing

MCP SecuritySeptember 29, 2026

PostgreSQL MCP Security: Database Permissions and Agent Access Control

Learn how to secure PostgreSQL MCP access with least-privilege roles, approved views, row-level security, export controls, and runtime authorization.

SH

Sachin HHead of Marketing

MCP SecuritySeptember 28, 2026

Salesforce MCP Security: Data Access, Permissions, and Agent Risk

Learn how to secure Salesforce MCP access with least-privilege OAuth scopes, object and field permissions, record-level controls, approval gates, and runtime authorization.

SH

Sachin HHead of Marketing

MCP SecuritySeptember 27, 2026

GitHub MCP Security Guide: Securing Agent Actions, Permissions, and Runtime Access

GitHub MCP can let agents read issues, inspect pull requests, edit files, run workflows, and merge code. This guide shows how to secure those actions with scoped permissions, tool restrictions, approval gates, and runtime authorization.

SH

Sachin HHead of Marketing

MCP SecuritySeptember 26, 2026

Slack MCP Security: OAuth Scopes, Bot Permissions, and Agent Access Control

Slack access is not simply read vs write. This guide explains how to secure Slack MCP with scopes, token design, channel boundaries, approvals, and checks before agent actions run.

SH

Sachin HHead of Marketing

MCP SecuritySeptember 25, 2026

MCP Server Security: A Production Hardening Guide

MCP servers expose tools, accept tool-call parameters, and broker access to downstream systems. This guide explains how to harden them before agents connect to production.

SH

Sachin HHead of Marketing

ConceptualSeptember 24, 2026

Human Approval Policies for AI Agents: Designing Risk-Based Authorization Workflows

A practical guide to deciding which AI agent actions can execute automatically, which require human approval, and how approved actions should be enforced.

SH

Sachin HHead of Marketing

ConceptualSeptember 23, 2026

Machine Identity Management for AI Agents: Inventory, Ownership, and Runtime Access

Learn how to manage AI agent identities, ownership, permissions, and runtime access before agents take consequential actions.

SH

Sachin HHead of Marketing

ConceptualSeptember 22, 2026

AI Agent Supply Chain Security: Assessing MCP Servers, Tools, and Third-Party Connectors

Review third-party MCP connectors, detect changing tools, and limit what AI agents can do with approved integrations.

SH

Sachin HHead of Marketing

MCP SecuritySeptember 21, 2026

MCP Security Audit: How to Assess Server Authentication, Tool Exposure, and Access Risks

What can your AI agents access through MCP? Learn how to inventory exposed tools, assess authentication, test authorization, and examine task-specific access.

SH

Sachin HHead of Marketing

ConceptualSeptember 20, 2026

Policy as Code for AI Agents: How to Define and Enforce Agent Authorization Policies

Define policy as code for AI agents, evaluate exact tool calls, and enforce allow, approval, or deny decisions before execution.

SH

Sachin HHead of Marketing

ConceptualSeptember 19, 2026

Continuous Access Evaluation vs Runtime Authorization for AI Agents

Continuous access evaluation communicates changes in security state. Runtime authorization determines whether an AI agent's next action is permitted. Learn how both mechanisms work together.

SH

Sachin HHead of Marketing

ConceptualSeptember 18, 2026

AI Agents vs. Service Accounts: Why Agents Need a Different Access Model

Service accounts identify workloads. AI agent identity distinguishes the agent operating within or through that workload. Learn when service accounts are enough and when agents need additional identity, delegation, and runtime authorization.

SH

Sachin HHead of Marketing

ConceptualSeptember 17, 2026

OAuth Scopes for AI Agents: What They Control and What They Don't

OAuth scopes define the capabilities available through an access token. For AI agents, least privilege may also require task-specific access decisions at execution time.

SH

Sachin HHead of Marketing

ConceptualSeptember 16, 2026

AI Agent Authentication vs. Authorization: What's the Difference?

Authentication verifies an AI agent's identity. Authorization determines what that identity can access or do. Runtime authorization evaluates a specific action when its context is known.

SH

Sachin HHead of Marketing

ConceptualSeptember 15, 2026

AI Agent Identity vs Workload Identity: What's Different?

Workload identity identifies the software runtime. AI agent identity identifies the agent operating within it. Learn where the two overlap, when workload identity is enough, and where runtime access control becomes important.

SH

Sachin HHead of Marketing

MCP SecuritySeptember 14, 2026

MCP Server Authentication: How Identity Works Across Clients, Users, and AI Agents

MCP authentication does not create one identity across an agent workflow. Learn how OAuth tokens establish identity context, what an MCP server can verify, and what happens when an agent reaches downstream services.

SH

Sachin HHead of Marketing

ConceptualSeptember 13, 2026

AI Agent Authentication Methods: API Keys, OAuth, Workload Identity, mTLS, and Short-Lived Credentials

Compare API keys, OAuth, workload identity, mTLS, SPIFFE, and short-lived credentials for authenticating AI agents and securing autonomous workflows.

SH

Sachin HHead of Marketing

ConceptualSeptember 12, 2026

Zero Standing Privilege for AI Agents: Why Privileged Access Should Exist Only for the Task

Learn how Zero Standing Privilege keeps privileged authority unavailable by default and grants AI agents task-scoped access only when work requires it.

SH

Sachin HHead of Marketing

ConceptualSeptember 11, 2026

Just-in-Time Access for AI Agents: How JIT Access Works

Learn how Just-in-Time access gives AI agents temporary, task-specific authority while reducing standing privilege and enforcing runtime access control.

SH

Sachin HHead of Marketing

ConceptualSeptember 10, 2026

RBAC vs ABAC for AI Agents: Which Authorization Model Fits Agentic Systems?

RBAC sets the baseline boundary for AI agents, while ABAC evaluates whether a specific tool action is authorized in its runtime context.

SH

Sachin HHead of Marketing

ConceptualSeptember 9, 2026

Top AI Agent Identity Management and IAM Platforms in 2026

Compare 10 AI agent identity management and IAM platforms for identity, governance, non-human identity security, credentials, delegated access, and runtime authorization.

SH

Sachin HHead of Marketing

ConceptualSeptember 6, 2026

Runtime Access Control for AI Agents: Why Permissions Need to Be Enforced at Execution Time

Runtime access control for AI agents determines whether a specific agent action should be allowed when the action is attempted and before it executes.

SH

Sachin HHead of Marketing

ConceptualSeptember 5, 2026

Least Privilege for AI Agents: How to Enforce It in Production

A practical security playbook for controlling agent identities, tools, actions, resources, delegated access, and runtime authorization.

SH

Sachin HHead of Marketing

ConceptualSeptember 4, 2026

AI Agent Permissions: How Enterprises Can Control What Agents Access and Do

AI agents are moving from controlled pilots into real enterprise workflows. As they begin accessing tools, data, APIs, and business systems, permissions become core enterprise control.

SH

Sachin HHead of Marketing

ConceptualSeptember 3, 2026

AI Agent Authorization: How to Control What Agents Can Do

How to manage agent permissions, apply least privilege, and control tool and API actions at runtime.

SH

Sachin HHead of Marketing

MCP SecuritySeptember 2, 2026

MCP Security Best Practices for Production Deployments

MCP ships no mandatory security. Here's the practitioner checklist: OAuth 2.1 + PKCE, per-tool authorization, vault-injected credentials, tool call validation, and tamper-evident audit logs.

SK

Sundar KrishFounder and CEO

MCP SecuritySeptember 1, 2026

MCP Gateway vs Embedded Proxy: Where Should Tool-Call Enforcement Run?

MCP has two deployment models: local STDIO and remote gateways. Neither enforces policy at the tool-call level inside your trust boundary. Here's the architecture that does.

SK

Sundar KrishFounder and CEO

TechnicalJuly 16, 2026

Why Multi-Step Drift Isn't Required for Runtime Risk

We attempted to induce multi-step MCP agent drift across monitoring and AWS tools. The experiment didn't produce reliable chains—but it revealed why the very first unexpected tool call is already a runtime security decision.

SK

Sundar KrishFounder and CEO

ConceptualJune 22, 2026

Four Ways AI Agents Mishandle Credentials Today, and Why Each Creates an Unexpectedly Broad Blast Radius

AI agents inherit credential patterns built for humans and static workloads. Each anti-pattern looks manageable in isolation. Together, they create a blast radius that covers everything the credential can touch.

SK

Sundar KrishFounder and CEO

ConceptualMay 29, 2026

Tracing Access from Human Intent to Tool Execution: Why Every AI Agent Tool Call Needs a Complete Delegation Chain Record

Every AI agent tool call is the end of a chain — from human intent through agent reasoning to execution. Without a record of how authority traveled that chain, authorized and unauthorized look identical after the fact.

SK

Sundar KrishFounder and CEO

ConceptualMay 28, 2026

Observability for AI Agents Starts Inside the Enforcement Layer, Not Outside It

Standard monitoring tools capture what your agents did — not whether they were allowed to do it. Here's why security observability for AI agents must start inside the enforcement layer.

SK

Sundar KrishFounder and CEO

TechnicalMay 26, 2026

I Told the Orchestrator 'Do Not Modify.' 7 of 12 Sub-agents Never Heard It.

What 36 controlled runs of Claude Code's Agent primitive reveal about delegation: the user's no-modify constraint disappears at the sub-agent boundary 7 times in 12, and the sub-agent that doesn't hear it inherits 41 GitHub tools plus Bash, Edit, and Write.

SK

Sundar KrishFounder and CEO

MCP SecurityMay 13, 2026

MCP's Missing Authorization Layer: Why Tool Discovery and Access Control Are Not the Same Thing

MCP handles tool discovery — not what happens when an agent calls one. The authorization gap that OAuth doesn't close requires enforcement at execution time.

SK

Sundar KrishFounder and CEO

ConceptualMay 7, 2026

AI Agents Are Not Your Average Identity: Here's Why They Need Their Own System

AI agents are not service accounts or bots. They're a third identity category: non-deterministic, autonomous, and built to chain tools at machine speed. Here's what that means for IAM.

SK

Sundar KrishFounder and CEO

ConceptualMay 1, 2026

Why AI Agent Access Control Must Happen at Execution Time, Not at Configuration Time

Static permissions assigned to AI agents at configuration time leave gaps at every tool call. This article explains why runtime access control with scoped, ephemeral credentials is essential for AI agents.

SK

Sundar KrishFounder and CEO

TechnicalApril 30, 2026

I Gave My LangChain Agent Three Tools. All Three Had Access to Every Secret in the Process.

A real LangChain setup showed how Slack, SQL, and web search tools can all access the same shared secrets when there is no runtime isolation.

SK

Sundar KrishFounder and CEO

TechnicalApril 29, 2026

How a Tool Description Poisoning Attack Sent My Agent Out of Scope

How one poisoned tool description pushed an agent out of scope and into AWS

SK

Sundar KrishFounder and CEO

ConceptualApril 20, 2026

Why Traditional IAM Fails When an AI Agent Calls a Tool

AI agents violate every assumption traditional IAM was built on. This article examines four specific failures at the tool-call layer — and what runtime enforcement needs to look like when agents act at machine speed.

SK

Sundar KrishFounder and CEO

TechnicalApril 15, 2026

GitHub MCP Gave My Agent 41 Tools. A 3-Word Prompt Made It Use the Wrong Ones.

How vague prompts turn read-only tasks into write attempts — and why static token scoping isn't enough to stop it.

SK

Sundar KrishFounder and CEO

TechnicalApril 9, 2026

My Agent Finished Its Job, Then Started another one out-of-scope

How a simple monitoring task drifted into AWS infrastructure

SK

Sundar KrishFounder and CEO