What is Postman?
Postman is an API engineering platform that gives AI agents structured context, skills, and governance to build, test, and operate APIs. Postman differentiates itself through the Context Graph, a queryable map of an organization's APIs, collections, and relationships that grounds agent-generated calls and tests in how APIs actually work.
Key Features
- Context Graph: A structured, queryable map of APIs, collections, and their relationships that grounds agent-generated calls and tests in real API behavior. Postman reports that an agent using the Context Graph found 498 services in a checkout scenario versus 129 with file search alone.
- Agent skills via Postman CLI: Installable with
npm install -g postman-cliand initialized withpostman init, these skills cover api-discovery, api-testing, api-mocking, api-monitoring, ci-integration, and performance-testing. - API discovery: Finds and reuses APIs that already exist within an organization, so agents and developers do not rebuild what is already available.
- API testing: Generates request collections and assertions designed to catch regressions.
- API mocking: Spins up a mock server before the real API exists, allowing work to proceed against a simulated endpoint.
- API monitoring: Watches live endpoints for uptime and drift.
- CI integration: Runs collection and governance checks on every commit.
- Performance testing: Simulates real-world load to surface bottlenecks.
Use Cases
- Agent-assisted API development: A developer prompts an agent to create a collection for the API in a repository, add tests, and run them, with the Context Graph supplying the grounding context.
- Change impact analysis: Before modifying a service such as a credit card service, a team asks which company APIs use it, or what depends on a billing-api collection and what a response schema change would break.
- Pre-release validation: Teams run collection and governance checks in CI on every commit and simulate load with performance testing to catch regressions and bottlenecks before production.
- Early-stage API design: Developers spin up a mock server before the API exists so dependent work can continue.
- Production monitoring: Teams watch live endpoints for uptime and drift after deployment.
Who It Is For
Postman targets developers, API teams, and platform engineers who want AI agents to work accurately against real APIs rather than guessing. It suits organizations that need agents to discover existing services, generate grounded tests, and operate within governance controls, and it is backed by a user base the source describes as over 50 million developers and 500,000 companies.
Pricing
The source does not state pricing figures. It offers a free sign-up option ("Sign up for free") and a desktop install path alongside the CLI.
Frequently Asked Questions
What is Postman used for?
Postman is used for API engineering, including discovering existing APIs, generating request collections and assertions, mocking APIs before they exist, monitoring live endpoints, running governance checks in CI, and performance testing. It also provides AI agents with structured context about how an organization's APIs work.
How do AI agents use Postman?
Agents use Postman through installable skills delivered by the Postman CLI (npm install -g postman-cli, then postman init). These skills let agents discover APIs, generate and run tests, create mock servers, monitor endpoints, and run CI and performance checks, with the Context Graph grounding their output.
What is the Postman Context Graph?
The Context Graph is a structured, queryable map of an organization's APIs, collections, and relationships. It gives agents full-system visibility so generated calls and tests reflect actual API behavior; Postman reports an agent found 498 services in a checkout with the Context Graph versus 129 without it.
Is Postman free?
The source lists a free sign-up option and does not publish pricing figures. Teams should check the Postman website for current plan details.
Who should use Postman?
Postman is aimed at developers and API teams who want AI agents to build and test against real APIs with fewer tokens and higher accuracy. It also fits organizations that need governance, credential control, and CI checks around API usage.
