Back to Devoured - September 21, 2026
DEVOURED
PGRun (Tool)

PGRun (Tool)

Data databasepostgresqldevopscloud pgrun.dev
PGRun enables developers and CI systems to spin up isolated, ephemeral Postgres branches from production without managing manual staging infrastructure.
What: PGRun allows users to connect an existing Postgres production database and programmatically create disposable branches for agents, CI pipelines, or pull requests. It provides a standard `DATABASE_URL` and ensures that sensitive production data can be masked during branch creation.
Why it matters: This addresses the bottleneck of shared staging environments where multiple concurrent tests or AI agents cause migration collisions and state corruption.
Takeaway: Register at pgrun.dev and use `pgrun branch create --from production` to isolate your agent's test environment.
Deep dive
  • PGRun acts as an infrastructure runtime, not just a dashboard.
  • Branches are ephemeral, designed to be deleted after CI tasks finish, leaving no state behind.
  • No custom ORM or query language is required; it uses standard Postgres connectivity.
  • It solves the 'shared staging' problem by providing a unique database per agent or pull request.
Decoder
  • Ephemeral: Infrastructure or resources that exist only for the duration of a task and are discarded afterward.
  • Staging: A testing environment that is intended to be a mirror of production.
  • Migration: A script or file used to update the database schema version.
Original article

Your production database stays where it is.

Connect your existing Postgres once. Then agents, CI, and pull requests can create isolated, disposable branches on demand.

One staging database doesn't work for 20 agents.

When multiple agents share the same staging database, migrations collide, tests interfere with each other, and state becomes unpredictable. With PGRun, every agent gets its own isolated Postgres.

agent-1 ─┐
agent-2 ─┼→ staging
agent-3 ─┘

Migrations collide. Tests share state. Cleanup becomes another job.

agent-1 → branch-1
agent-2 → branch-2
agent-3 → branch-3

No collisions. No shared state. Delete everything when the work is done.

Your agents never touch production.

Connect production once. PGRun creates isolated, disposable Postgres branches for agents, CI, and pull requests.

Production credentials never reach the agent.

Sensitive data can be detected and masked before it reaches a branch.

Every branch is isolated and disposable.

Postgres runtime for agents

Give agents infrastructure they can create, use, and destroy programmatically — without another database dashboard.

Connect

Connect your existing Postgres — production stays put.

Branch

Spin up an isolated Postgres branch in seconds.

Run

Get a normal DATABASE_URL. No custom API.

Destroy

Delete it when the job's done. No cleanup.

Built for agents

CLI, API, and structured output for agents.

It's just Postgres.

No custom database API. No proprietary query language. No special ORM. Every branch gives you a normal:

DATABASE_URL=postgres://...

Why pgrun

Not just a dashboard. A runtime.

pgrun gives agents infrastructure they can create, use, and destroy on their own.

Not just for one database.

Every agent, PR, and CI run gets its own isolated branch, in parallel.

Not just fast to create.

Branches are cheap to delete too — no leftover state, no cleanup jobs.

Not another database API.

A normal DATABASE_URL — works with anything that speaks Postgres.

Start with $10 free credits.

No credit card required. Create real Postgres branches, connect your agents, and test the workflow before paying.

Point it at a database.

Start in the terminal

$ pgrun login
$ pgrun source add production
$ pgrun branch create agent-42 --from production

Or create branches directly through the API.

Explore all 67 articles in today's digest

Fresh dev news, AI-digested daily. Never miss a trend at devoured.fyi