name: guidewire-local-dev-loop description: Iterate on Gosu rules and configuration without paying the full 5–15 minute runServer rebuild every time. Use when standing up Guidewire Studio against a local InsuranceSuite instance, attaching an IntelliJ remote debugger to runServer, distinguishing changes that hot-reload from changes that force restart, or building a GUnit-driven TDD cycle for rule logic. Trigger with "guidewire studio", "gosu hot reload", "gosu debugger", "gunit", "guidewire runServer". allowed-tools: Read, Write, Edit, Bash(gradle:), Bash(java:), Bash(jdb:*), Grep, Glob version: 2.0.0 license: MIT author: Jeremy Longshore jeremy@intentsolutions.io compatibility: Designed for Claude Code, also compatible with Codex and OpenClaw tags:

• guidewire
• gosu
• studio
• hot-reload
• gunit
• debugging

Guidewire Local Dev Loop

Overview

Run a local InsuranceSuite instance and iterate on Gosu rule logic in seconds, not minutes. The single biggest productivity killer in Guidewire development is paying the 5–15 minute gradle runServer cold-start cost on every change because the developer does not know which edits hot-reload and which force a restart.

Three production problems this skill prevents:

1. Restart cascade — developer changes a Gosu rule, restarts runServer, waits 8 minutes, finds the rule was wrong, repeats. A full day disappears in restarts.
2. Silent stale code — Studio claims it hot-reloaded a class but the running JVM is still executing the old bytecode (common when interfaces change). Tests pass against stale code.
3. GUnit drift — unit tests for Gosu rules diverge from the rules themselves because the cycle to run a single test through Studio is too slow; developers stop writing them.

Prerequisites

• JDK 17 (for Cloud release 202503+)
• Guidewire Studio installed (IntelliJ-based, distributed by Guidewire)
• Local InsuranceSuite configuration zone (PolicyCenter, ClaimCenter, or BillingCenter)
• ≥16 GB RAM on the dev machine — runServer + Studio + the JVM debug agent need headroom
• Sample data loader configured for the chosen product (e.g., PersonalAuto for PC)

Instructions

Build the inner loop in this order. Every step targets one of the three productivity killers above.

1. Start runServer once, keep it warm

Cold start takes 5–15 minutes; treat it as a session investment.

# Start in dev mode with debug agent on 8088, leaves the server attached to the terminal
./gradlew runServer -Pdebug=true -PdebugPort=8088 -Dgw.servermode=dev

gw.servermode=dev enables the hot-reload paths inside the JVM. debugPort=8088 exposes the JDWP debug agent — attach IntelliJ to it once and leave it. Restart only when the what hot-reloads table below says you must.

2. What hot-reloads, what does not

Memorize this table — it determines whether the next edit costs 0 seconds or 8 minutes.

When in doubt, trust the JVM, not Studio. Open the IntelliJ debugger, set a breakpoint on the changed method, trigger the code path, and confirm the breakpoint hits the new line numbers. Studio's "reloaded" status is informational, not authoritative.

3. Attach the IntelliJ debugger once per session

Run > Edit Configurations > + > Remote JVM Debug
  Host: localhost
  Port: 8088
  Module classpath: <your-config-module>
  Save → run with the bug icon

Once attached, breakpoints survive Gosu hot-reloads. The connection drops only on full runServer restart. Use conditional breakpoints (policy.totalPremium.compareTo(BigDecimal("10000")) > 0) for production-shaped data — never trust toy values.

4. GUnit cycle for rule TDD

Gosu rules are testable without a running server. GUnit tests run in seconds and should drive every non-trivial rule change.

# Run a single GUnit test class
./gradlew test --tests "com.acme.policycenter.rules.UnderwritingIssueRuleTest"

# Run all rule tests in a package, with continuous re-run on change
./gradlew test --tests "com.acme.policycenter.rules.*" --continuous

--continuous reruns the matching tests every time a file changes. Pair with the rule under test in a split editor — feedback loop drops to <5 seconds per save.

5. Sample data isolation per session

Every developer needs a deterministic fixture set, not whatever junk is in the shared dev database. Load a per-session sample at runServer start:

# Load the standard sample, then a project-specific overlay
./gradlew loadSampleData -PsampleData=default -PsampleData=acme-uat-fixtures runServer

Project-specific sample sets live in modules/configuration/test/data/ and are checked in. Treat the dev database as ephemeral — never store work-in-progress data only in it; it dies on the next dropAndCreateDatabase.

Output

A working local dev loop ships with all of the following:

• gradle runServer running in dev mode with debugPort=8088 exposed; remote-debug connection attached from IntelliJ.
• The hot-reload-vs-restart table internalized and applied — at least 80% of edits cost zero restart time.
• A gradle test --continuous watcher running in a side terminal for GUnit-driven TDD on the current rule.
• Sample data loaded from a checked-in fixture set, reproducible across team members.
• A breakpoint validation habit: every non-trivial rule change is confirmed hot-reloaded by hitting a breakpoint in the new line, not by trusting Studio's reload indicator.

Examples

Example 1 — Pure rule edit, zero restart

1. Edit rule body in modules/configuration/gsrc/.../UnderwritingIssueRules.gs
2. Save (Ctrl-S)
3. Trigger the rule (issue a quote in the running PC instance)
4. Breakpoint hits the new line numbers; rule fires with new logic
5. Total time from save to confirmation: <10 seconds

Example 2 — Interface change, controlled restart

1. Modify interface in modules/configuration/gsrc/.../IPolicyCalculator.gs
2. Recognize this is in the no-hot-reload row of the table
3. Stop runServer (Ctrl-C); start ./gradlew runServer -Pdebug=true -PdebugPort=8088
4. Wait ~8 minutes; reattach debugger
5. Resume work — accept the cost rather than chasing phantom bugs from stale bytecode

Example 3 — TDD cycle on a new validation rule

# Terminal 1: continuous test runner
./gradlew test --tests "com.acme.policycenter.validation.HighValueAccountValidatorTest" --continuous

# Editor: write the failing test first, watch it fail in <5s
# Implement the rule, watch the test pass in <5s
# Commit when green; do not run the full server until the rule is locked

Error Handling

For deeper coverage (containerized dev environments, multi-developer shared servers, plugin debug logging, custom datasource hooks), see implementation guide and API reference.

See Also

• guidewire-install-auth — once your local runServer integrates outbound to a Cloud tenant, auth layer applies the same as production
• guidewire-sdk-patterns — when local code calls Cloud API, the same client patterns apply
• guidewire-ci-cd-pipeline — promotion of locally-developed config through GCC slots; GUnit gates run there too
• guidewire-core-workflow-a — PolicyCenter workflows that local dev cycles target

Resources

• Guidewire Developer Portal
• Gosu language reference
• JDWP — Java Debug Wire Protocol
• Gradle continuous build