---
schema: "swft.publication/v1"
id: "company-spotify-honk"
title: "Spotify Honk: adding an agent to a software factory that already worked"
description: "How Spotify Honk combines Fleet Management, Backstage context, restricted tools, agent verification, an independent judge, sandboxes, and review."
summary: "Spotify already had rule-based systems for finding code repositories, applying repeated updates, opening review requests, and automatically merging well-understood maintenance work. Honk adds agent judgment where fixed scripts are not flexible enough, while the existing targeting, permissions, checks, and review stay in place."
canonical: "https://swft.io/companies/spotify-honk"
author: "SWFT Editorial"
author_type: "Organization"
published: "2026-09-01"
modified: "2026-09-02"
kind: "case-study"
section: "Companies"
tags: ["Spotify Honk", "background coding agent", "AI code migration", "deterministic automation", "software factory"]
evidence_labels: ["INFERENCE", "OBS", "SELF-REPORT"]
source_ids: ["spotify-code-no-longer-constraint", "spotify-fleetshift-product", "spotify-honk-migrations", "spotify-honk-part-one", "spotify-honk-part-three", "spotify-honk-part-two"]
authorship_disclosure: "AI-drafted from the cited public sources and independently checked by a second AI editorial-review agent (Codex) for source fit, claim boundaries, overlap, and reader utility. SWFT Editorial is responsible for corrections."
---

# Spotify Honk: adding an agent to a software factory that already worked

Spotify kept existing systems for choosing repositories, coordinating work, checking results, and merging safe changes, then used Honk for complex transformations.

> **Authorship:** AI-drafted from the cited public sources and independently checked by a second AI editorial-review agent (Codex) for source fit, claim boundaries, overlap, and reader utility. SWFT Editorial is responsible for corrections.

## Quick answer

Spotify already had rule-based systems for finding code repositories, applying repeated updates, opening review requests, and automatically merging well-understood maintenance work. Honk adds agent judgment where fixed scripts are not flexible enough, while the existing targeting, permissions, checks, and review stay in place.

Spotify's Honk case shows an important path into agentic software development. The company already had a rule-based system that found repositories, applied known migrations, opened pull requests, and merged safe changes. Honk adds agent judgment only where fixed transformation rules stop working.

## What is Spotify Honk?

Honk was developed as Spotify's internal background coding agent for large changes across many code repositories. A developer can describe a change in natural language and trigger it through Slack or GitHub. Honk receives the target repositories, works in an isolated environment, verifies the result, and prepares pull requests for review. Spotify now also offers [Honk-powered Fleetshift](https://backstage.spotify.com/fleetshift) as a managed product in Spotify Portal; Honk is not presented as open source.

Spotify [reported more than 1,500 merged AI-generated pull requests](https://engineering.atspotify.com/2025/11/spotifys-background-coding-agent-part-1) by November 2025. A later [dataset-migration case](https://engineering.atspotify.com/2026/4/background-coding-agents-dataset-migrations-honk-part-4) targeted about 1,800 direct pipelines and several thousand indirect dependents. Honk produced 240 automated pull requests; one unsupported framework was excluded, and owners had to test changes where unit tests were missing. Spotify estimated that the effort saved ten engineer-weeks. These are first-party project estimates rather than controlled productivity measurements.

## What existed before Honk?

Spotify's Fleet Management platform already knew which services and repositories the company owned. Fleetshift could apply deterministic code migrations, meaning the same fixed rules produced the same kind of change across the fleet. Spotify says this older system had produced [more than 2.5 million automated maintenance pull requests](https://engineering.atspotify.com/2026/6/code-with-claude-coding-is-no-longer-the-constraint), most of them merged automatically without human review.

That history matters. Target selection, orchestration, testing, pull-request creation, review policy, and merge rules were established before an agent entered the loop. Honk replaced the complex transformation step for work that was difficult to express as a fixed script.

The factory pattern is **deterministic rails with an agent in the uncertain middle**. Ordinary software selects and schedules the work. The agent reasons about each codebase. Verification may be automated or manual depending on repository support. Human review is required for some changes and safely omitted for established low-risk migrations.

## How does Spotify give Honk context and tools?

Spotify first experimented with several agent harnesses, the software shell that gives a model its instructions and tools, then moved to Claude Code. As described in [Part 2](https://engineering.atspotify.com/2025/11/context-engineering-background-coding-agents-part-2), the agent receives a small list of specifically approved tools: a verification interface, restricted source-control operations, and approved terminal commands. Prompts are static and versioned. Work is broken into one change at a time with explicit tests and end states.

For the large dataset migration, Backstage supplied service ownership and data lineage, code search found consumers, Fleetshift coordinated the targets, and Honk performed the code-specific transformation. Standardized repositories and good tests made success more likely.

## How does Honk verify its work?

Honk uses an inner loop that implements, verifies, and repairs. An independent model judge then decides whether the result is ready to continue. In [Part 3](https://engineering.atspotify.com/2025/12/feedback-loops-background-coding-agents-part-3), Spotify reported that the judge rejected roughly a quarter of sessions and that the agent recovered in about half of those rejected cases.

The company also stated an important limitation: it had not yet evaluated the judge itself rigorously. A model judge can catch weak work and can also approve or reject incorrectly. Its decision needs its own eval.

The agent runs in a sandbox with limited permissions. Git pushes and user interactions stay outside the agent boundary. These controls reduce the reach of a faulty run while preserving the ordinary review path.

## What has Spotify learned about the bottleneck?

By June 2026, Spotify [reported](https://engineering.atspotify.com/2026/6/code-with-claude-coding-is-no-longer-the-constraint) more than 99 percent weekly engineering use of AI tools, 94 percent self-reported productivity improvement among surveyed engineers, and a 76 percent increase in pull-request frequency. Those numbers combine several tools and practices. They are not Honk-only outcomes or an independent causal study.

Spotify's larger conclusion is more durable: coding stopped being the main constraint for some work. Review, product decisions, standardization, context quality, and coordination became more visible bottlenecks. A factory that only produces more changes can overload its acceptance system.

## What can a product team copy?

Find a deterministic maintenance line first. Keep repository discovery, targeting, scheduling, permissions, and merge policy in ordinary systems. Insert an agent only for the transformation that requires judgment. Give it limited tools, a clear end state, executable checks, an independent reviewer or judge, and a human escalation path.

Measure the complete migration: repositories attempted, changes accepted, failures, manual exceptions, review attention, defects, elapsed time, and customer or operational value. Pull-request volume alone cannot tell whether the factory improved the product.

## How we know

- **First-party report (SELF-REPORT)** Spotify describes Honk's origin, triggers, multi-repository work, merged-PR scale, harness, restricted tools, versioned prompts, and early eval limits. Sources: [Spotify's background coding agent, Part 1](https://engineering.atspotify.com/2025/11/spotifys-background-coding-agent-part-1); [Context engineering for background coding agents, Part 2](https://engineering.atspotify.com/2025/11/context-engineering-background-coding-agents-part-2).
- **First-party report (SELF-REPORT)** Spotify describes Honk's verifier and judge loop, sandbox boundary, judge limitation, and a production dataset migration using Backstage, code search, Fleetshift, and Honk. Sources: [Feedback loops for background coding agents, Part 3](https://engineering.atspotify.com/2025/12/feedback-loops-background-coding-agents-part-3); [Background coding agents for dataset migrations, Honk Part 4](https://engineering.atspotify.com/2026/4/background-coding-agents-dataset-migrations-honk-part-4).
- **First-party report (SELF-REPORT)** Spotify reports broad AI-tool adoption and output figures while identifying review, standardization, context, and decisions as emerging constraints. Sources: [Code with Claude: coding is no longer the constraint](https://engineering.atspotify.com/2026/6/code-with-claude-coding-is-no-longer-the-constraint).
- **Observed artifact (OBS)** Spotify offers Honk-powered Fleetshift as a managed product through Spotify Portal; the product surface does not present Honk as open source. Sources: [Fleetshift for Portal](https://backstage.spotify.com/fleetshift).
- **Analysis (INFERENCE)** SWFT treats deterministic rails with agent judgment in the uncertain middle as Spotify's most transferable pattern. Sources: [Background coding agents for dataset migrations, Honk Part 4](https://engineering.atspotify.com/2026/4/background-coding-agents-dataset-migrations-honk-part-4); [Code with Claude: coding is no longer the constraint](https://engineering.atspotify.com/2026/6/code-with-claude-coding-is-no-longer-the-constraint).

## Sources

- **First-party report (SELF-REPORT)** [Code with Claude: coding is no longer the constraint](https://engineering.atspotify.com/2026/6/code-with-claude-coding-is-no-longer-the-constraint) — Spotify Engineering; published 2026-06-03; accessed 2026-09-01. Spotify's current adoption account and its warning that review, standardization, context, and organizational decisions become the constraints as code generation accelerates.
- **Observed artifact (OBS)** [Fleetshift for Portal](https://backstage.spotify.com/fleetshift) — Spotify for Backstage; accessed 2026-09-01. Spotify's managed Fleetshift product page states that Honk powers agentic code changes across repositories; it does not present Honk as an open-source release.
- **First-party report (SELF-REPORT)** [Background coding agents for dataset migrations, Honk Part 4](https://engineering.atspotify.com/2026/4/background-coding-agents-dataset-migrations-honk-part-4) — Spotify Engineering; published 2026-04-22; accessed 2026-09-01. Spotify's production migration case connecting Backstage lineage, code search, deterministic targeting, agent transformation, and verification.
- **First-party report (SELF-REPORT)** [Spotify's background coding agent, Part 1](https://engineering.atspotify.com/2025/11/spotifys-background-coding-agent-part-1) — Spotify Engineering; published 2025-11-06; accessed 2026-09-01. Spotify's introduction to Honk, its Fleet Management ancestry, natural-language multi-repository work, human triggers, and early merged-PR scale.
- **First-party report (SELF-REPORT)** [Feedback loops for background coding agents, Part 3](https://engineering.atspotify.com/2025/12/feedback-loops-background-coding-agents-part-3) — Spotify Engineering; published 2025-12-09; accessed 2026-09-01. Spotify's description of an inner verifier loop, an independent judge, sandboxed execution, limited permissions, and judge-quality limitations.
- **First-party report (SELF-REPORT)** [Context engineering for background coding agents, Part 2](https://engineering.atspotify.com/2025/11/context-engineering-background-coding-agents-part-2) — Spotify Engineering; published 2025-11-24; accessed 2026-09-01. Spotify's account of harness choices, restricted tools, versioned prompts, task boundaries, verification, and immature structured evals.

## Read next

- [Block's AI software factory: Goose, Builderbot, and the Buzz workspace](/companies/block-builderbot-goose-buzz)
- [Stripe Minions: how developer infrastructure became an AI software factory](/companies/stripe-minions)
- [How an AI software factory works](/software-factory-architecture)