# Postman Blog The official Postman blog This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles. ## SDK vs CLI vs MCP, a dev story DevFeed: [SDK vs CLI vs MCP, a dev story]() Original publisher: [Read original article]() Author: Anthony Viard Published: 2026-09-17T16:00:00Z Content type: opinion Language: en Sources: [Postman Blog]() Topics: [SDKs](), [MCP Server](), [Command-line interface](), [GitHub API](), [Claude Code](), [GitHub]() Tags: [ai-agents](), [claude](), [cli](), [code](), [general](), [mcp](), [mcp-server](), [model-context-protocol](), [postman](), [sdk](), [sdk-generation]() ### AI overview A developer compares SDKs, command-line tools, and MCP servers as ways for an agent to access the GitHub API. Using an internal release-readiness tool as a case study, the article explains that SDKs suit scheduled, predefined code while interactive agent use may require more improvisation. ### Source excerpt SDK, CLI, or MCP server for your agent? I measured the token cost and auth tradeoffs of all three against one real API. See which to pick. The post SDK vs CLI vs MCP, a dev story appeared first on Postman Blog. ## Scheduling gRPC and GraphQL Requests with Postman Monitors (Beta) DevFeed: [Scheduling gRPC and GraphQL Requests with Postman Monitors (Beta)]() Original publisher: [Read original article]() Author: Harsh Vardhan Published: 2026-09-16T16:00:00Z Content type: tutorial Language: en Sources: [Postman Blog]() Topics: [gRPC](), [Postman](), [Monitoring](), [Authentication]() Tags: [beta](), [general](), [graphql](), [graphql-monitoring](), [how-to](), [monitor-grpc-streaming](), [monitoring](), [monitors](), [postman](), [postman-monitors-grpc-graphql](), [scheduling](), [test]() ### AI overview This tutorial explains how to schedule gRPC and GraphQL requests with Postman Monitors in beta. It covers supported request types, streaming caveats, response testing, authentication, and current feature limitations. ### Source excerpt Postman Monitors now run gRPC and GraphQL requests in beta. Learn how to schedule them, test responses, and send feedback. Start monitoring today. The post Scheduling gRPC and GraphQL Requests with Postman Monitors (Beta) appeared first on Postman Blog. ## Agent Gateway: The Next Evolution of the API Gateway DevFeed: [Agent Gateway: The Next Evolution of the API Gateway]() Original publisher: [Read original article]() Author: Gbadebo Bello Published: 2026-09-10T16:00:00Z Content type: article Language: en Sources: [Postman Blog]() Topics: [API](), [AI Agent](), [Architecture & Design](), [Artificial Intelligence](), [Routing (disambiguation)](), [Authentication](), [Authorization](), [observability]() Tags: [agent-gateway](), [ai](), [ai-agents](), [api](), [api-101](), [api-gateway](), [architecture](), [authorization](), [editor-s-picks](), [fabric-gateway](), [general](), [llm-gateway](), [mcp](), [mcp-gateway](), [observability](), [routing]() ### AI overview The article explains the emergence of the Agent Gateway as an architectural component for AI agents. It contrasts traditional API Gateways, which authenticate, govern, route, and deliver requests from clients that already know which API to call, with the needs of AI agents, which receive goals and may require broader governance across agentic workflows. ### Source excerpt API gateways are evolving for AI agents. Learn how the agent gateway governs identity, tools, memory, and policy across agentic workflows. Get early access. The post Agent Gateway: The Next Evolution of the API Gateway appeared first on Postman Blog. ## Loop engineering: stop prompting, start looping DevFeed: [Loop engineering: stop prompting, start looping]() Original publisher: [Read original article]() Author: Anthony Viard Published: 2026-09-08T16:00:00Z Content type: article Language: en Sources: [Postman Blog]() Topics: [Loop Engineering](), [AI Agent](), [AI-assisted coding](), [API]() Tags: [2026](), [agent](), [agentic](), [agents](), [ai](), [ai-agents](), [ai-coding](), [api](), [api-testing](), [blog](), [code](), [coding-agents](), [developer](), [engineering](), [general](), [guide](), [idea](), [loops](), [model](), [post](), [postman-cli](), [tool](), [verify](), [work]() ### AI overview This article explains loop engineering as a way to improve AI coding agent reliability. Instead of relying on a single prompt, the system repeatedly generates code, runs it against a real API or other source of truth, verifies the result, and decides whether to stop or continue. The article distinguishes this outer verification loop from the inner ReAct tool-use cycle and presents a reproducible setup for allowing an agent to correct its own mistakes. ### Source excerpt Stop prompting your AI agent, start looping. A guide to loop engineering: wire a real API call in as the oracle so generated code stops guessing. The post Loop engineering: stop prompting, start looping appeared first on Postman Blog. ## Catching Salesforce Custom Field Drift With the Postman AI Engineer DevFeed: [Catching Salesforce Custom Field Drift With the Postman AI Engineer]() Original publisher: [Read original article]() Author: Talia Kohan Published: 2026-09-03T16:00:00Z Content type: article Language: en Sources: [Postman Blog]() Topics: [Postman](), [Claude](), [Automation](), [API](), [Front end](), [OpenAPI Specification](), [GitHub](), [Back end](), [Pull Request](), [payload](), [Slack]() Tags: [ai](), [ai-automation](), [api](), [api-drift](), [backend](), [code](), [frontend](), [general](), [github](), [payload](), [postman-agent-mode](), [pull-request](), [salesforce](), [slack]() ### AI overview This Postman Blog article explains how a Salesforce custom field caused drift between an API collection, frontend code, and an OpenAPI specification. It describes using the Postman AI Engineer to diagnose the failing request and Postman AI Automation to update the affected artifacts and add tests through a GitHub pull request. ### Source excerpt A Salesforce custom field broke my app. See how Postman Agent Mode found the drift and how a Skill kept it from happening again. The post Catching Salesforce Custom Field Drift With the Postman AI Engineer appeared first on Postman Blog. ## How AI Agents Discover and Integrate Public APIs DevFeed: [How AI Agents Discover and Integrate Public APIs]() Original publisher: [Read original article]() Author: Anthony Viard Published: 2026-09-02T16:00:00Z Content type: article Language: en Sources: [Postman Blog]() Topics: [AI Agent](), [API](), [Postman](), [MCP]() Tags: [ai-agent](), [ai-agents](), [api](), [api-discovery](), [apis](), [general](), [integration](), [mcp](), [model-context-protocol](), [orbit]() ### AI overview The article presents Orbit by Postman, which lets developers and AI agents discover public API endpoints by describing a desired workflow in plain language. Orbit evaluates endpoint capabilities, identifies gaps, and generates integration details such as authentication, URLs, parameters, response shapes, execution order, and implementation pitfalls. It can be accessed through search and integration endpoints or via the Model Context Protocol (MCP). ### Source excerpt I couldn't name the API my feature needed. Orbit found it in one search, then handed me the request bodies and the three bugs I'd have shipped. The post How AI Agents Discover and Integrate Public APIs appeared first on Postman Blog. ## Finding three-way API drift with the AI Engineer DevFeed: [Finding three-way API drift with the AI Engineer]() Original publisher: [Read original article]() Author: Talia Kohan Published: 2026-09-01T16:00:00Z Content type: article Language: en Sources: [Postman Blog]() Topics: [API](), [Postman](), [OpenAPI Specification](), [Claude](), [Tooling](), [client](), [servers]() Tags: [ai-engineer](), [api](), [api-drift](), [apis](), [ci](), [fintech](), [general](), [openapi](), [postman-ai-engineer](), [production](), [qa]() ### AI overview The article explains how to detect three-way API drift by comparing an OpenAPI specification, the requests in Postman Collections, and the responses returned by a running server. It argues that pairwise checks can miss cases where all three sources disagree, and describes using Postman AI Engineer and the Context Graph to identify these inconsistencies. ### Source excerpt Find three-way API drift with Postman AI Engineer and the Context Graph. Compares your OpenAPI spec, collection requests, and live server responses. The post Finding three-way API drift with the AI Engineer appeared first on Postman Blog. ## QE Platform Strategy: 3 Paths and Where Each Hides Its Cost DevFeed: [QE Platform Strategy: 3 Paths and Where Each Hides Its Cost]() Original publisher: [Read original article]() Author: Rick Crawford Published: 2026-08-31T17:37:05Z Content type: opinion Language: en Sources: [Postman Blog]() Topics: [Testing](), [Software Engineering](), [Development](), [Artificial Intelligence](), [Open Source]() Tags: [ai](), [api-testing](), [blog](), [development](), [general](), [open-source](), [qe-program](), [quality-engineering](), [sdlc](), [software-engineering](), [strategy](), [testing]() ### AI overview This opinion article compares three QE platform strategies: continuing with the current approach, building an internal platform, or integrating specialized tools. It explains the hidden costs of each path, including test maintenance, delayed releases, escaped defects, long-term platform maintenance, dependency updates, security patches, and the need to modernize for AI-assisted testing. ### Source excerpt In Part 1 I named the six structural problems compounding inside QE programs as AI accelerates development. In Part 2 I walked... The post QE Platform Strategy: 3 Paths and Where Each Hides Its Cost appeared first on Postman Blog. ## Introducing Orbit: Turn Any Task Into the Right API Calls DevFeed: [Introducing Orbit: Turn Any Task Into the Right API Calls]() Original publisher: [Read original article]() Author: Abhinav Asthana Published: 2026-08-27T16:00:00Z Content type: article Language: en Sources: [Postman Blog]() Topics: [API](), [MCP Server](), [Artificial Intelligence](), [MCP](), [Postman](), [REST API](), [Claude Code]() Tags: [agents](), [ai](), [ai-agents](), [api-discovery](), [blog](), [free](), [general](), [mcp](), [mcp-server](), [model-context-protocol](), [orbit](), [post](), [rest-api]() ### AI overview Postman introduces Orbit, a free API discovery tool for AI agents. Agents can describe a task to find suitable public API endpoints, then select an endpoint and receive an integration brief covering authentication, URLs, request steps, parameters, response codes, dependencies, and common pitfalls. Orbit is available as an MCP server and a REST API, with no account, API key, or OAuth setup required. ### Source excerpt Orbit is a free API discovery tool for AI agents. Describe a task, get the endpoints that fit, then get a task brief to integrate them. No account, no auth. The post Introducing Orbit: Turn Any Task Into the Right API Calls appeared first on Postman Blog. ## Why AI coding agents need context graphs DevFeed: [Why AI coding agents need context graphs]() Original publisher: [Read original article]() Author: Talia Kohan Published: 2026-08-25T16:00:00Z Content type: article Language: en Sources: [Postman Blog]() Topics: [AI-assisted coding](), [Graphs](), [coding](), [API](), [internal developer platform](), [context window](), [long-context]() Tags: [ai-coding-agents](), [api-governance](), [apis](), [context-window](), [developer](), [developer-platform](), [engineering](), [general](), [graphs](), [knowledge-graphs]() ### AI overview The article argues that AI coding agents struggle in real codebases primarily because they lack access to a knowledge graph connecting code, APIs, services, ownership, dependencies, policies, and related organizational context. It describes service catalogs, API registries, ownership maps, dependency graphs, and internal developer platforms as different forms of the same underlying structure. It also argues that simply increasing the context window does not solve retrieval and context-quality problems. ### Source excerpt AI coding agents don't fail from small context windows. They fail without a knowledge graph of your code, APIs, and vendors. The post Why AI coding agents need context graphs appeared first on Postman Blog. ## How Postman Passport keeps API secrets inside your network DevFeed: [How Postman Passport keeps API secrets inside your network]() Original publisher: [Read original article]() Author: Talia Kohan Published: 2026-08-24T16:00:00Z Content type: article Language: en Sources: [Postman Blog]() Topics: [API](), [Postman](), [API keys](), [Security](), [Architecture & Design](), [Network]() Tags: [api](), [api-keys](), [api-security](), [architecture](), [general](), [network](), [postman-passport](), [secret-store](), [secrets](), [secrets-management](), [security](), [token]() ### AI overview Postman Passport keeps real API secrets inside an organization's network by issuing credential references instead of exposing underlying keys. An internal proxy resolves the references, injects the credentials, and forwards requests while cryptographically binding references to their authorized holders. ### Source excerpt Postman Passport grants API access using credential references, keeping the real secret inside your network. Here's how the architecture works. The post How Postman Passport keeps API secrets inside your network appeared first on Postman Blog. ## 5 QE Pipeline Metrics That Show Where Quality Is Leaking DevFeed: [5 QE Pipeline Metrics That Show Where Quality Is Leaking]() Original publisher: [Read original article]() Author: Rick Crawford Published: 2026-08-21T15:00:10Z Content type: article Language: en Sources: [Postman Blog]() Topics: [Development](), [Postman]() Tags: [api-testing](), [developer](), [development](), [false-positives](), [general](), [metrics](), [packages](), [policy](), [production](), [qe-program](), [quality-engineering](), [sdlc](), [shift-left](), [validation]() ### AI overview This developer article presents five quality-engineering pipeline metrics aligned with Design, Gate, Validate, Monitor, and Improve stages. The supplied text details package adoption, gate coverage and flake rate, and defect escape rate, explaining how they reveal uneven coverage, ineffective merge gates, false positives, and defects reaching production. ### Source excerpt In the last post I described six problems that compound when development outruns the QE infrastructure underneath it. Most leaders recognize all... The post 5 QE Pipeline Metrics That Show Where Quality Is Leaking appeared first on Postman Blog. ## The prompts only the Postman AI Engineer can answer DevFeed: [The prompts only the Postman AI Engineer can answer]() Original publisher: [Read original article]() Author: Talia Kohan Published: 2026-08-20T16:00:00Z Content type: article Language: en Sources: [Postman Blog]() Topics: [Claude](), [Postman](), [Graphs](), [API](), [OpenAPI Specification](), [monitor](), [React]() Tags: [ai](), [ai-agents](), [api](), [apis](), [general](), [monitor](), [openapi](), [postman-ai-engineer](), [react]() ### AI overview This Postman Blog article presents the Postman AI Engineer as a natural-language interface over the Postman Context Graph. It describes prompts that trace identifiers across APIs, collections, runtime monitor results, repositories, and React dashboards, as well as prompts that expose specification-versus-code drift and three-way discrepancies involving live server responses. ### Source excerpt The Postman AI Engineer prompts that trace an ID across your APIs, catch three-way drift, and auto-write a PR review from the graph. The post The prompts only the Postman AI Engineer can answer appeared first on Postman Blog. ## Develop and test MCP in Postman: jhipster-mcp in action DevFeed: [Develop and test MCP in Postman: jhipster-mcp in action]() Original publisher: [Read original article]() Author: Anthony Viard Published: 2026-08-19T16:00:00Z Content type: article Language: en Sources: [Postman Blog]() Topics: [Model Context Protocol](), [Postman](), [Spring Boot](), [Node.js](), [TypeScript](), [Angular](), [React](), [Vue.js](), [JSON](), [Remote Procedure Call (RPC)]() Tags: [ai-agents](), [angular](), [apache](), [cli](), [general](), [jhipster](), [json](), [mcp](), [node-js](), [open-source](), [post](), [react](), [rpc](), [spring-boot](), [typescript](), [vue]() ### AI overview This Postman developer article demonstrates jhipster-mcp, an open-source MCP server that exposes the JHipster generator to MCP clients. It uses Postman's MCP client to test tools, prompts, and resources for generating and evolving Spring Boot applications from JDL. ### Source excerpt Drive the jhipster-mcp server from Postman's MCP client to test tools, prompts, and resources before handing them to an AI agent. Try it today. The post Develop and test MCP in Postman: jhipster-mcp in action appeared first on Postman Blog.