Browse reusable Agent Skills, each with a clear purpose and practical guidance.
Setting up a dedicated Golem environment for integration testing. Use when adding integration tests for a Golem application, configuring a throwaway local server for tests, isolating test state from local development, or wiring up CI to run end-to-end tests against a real Golem server.
Using the Golem REPL for interactive testing and scripting of agents. Use when asked to test agents interactively, run a REPL, or execute test scripts against deployed agents.
Run and debug Golem Scala SDK integration tests. Use when running golem-scala integration tests, debugging test failures, or working with GolemExamplesIntegrationSpec.
Simulating a crash on an agent for testing crash recovery. Use when testing durable execution, verifying state recovery after failures, or validating that an agent correctly resumes from its operation log.
Troubleshooting Golem build failures and debugging manifest file (golem.yaml) configuration. Use when a build fails, when environment variables or config values are not applied as expected, when diagnosing why a manifest property has the wrong value, or when using `golem component manifest-trace` to inspect how golem.yaml layers are resolved.
Investigating benchmark performance by running Golem benchmarks with OTLP tracing enabled and analyzing trace data from Jaeger. Use when diagnosing slow benchmarks, understanding benchmark call flows, or profiling span durations during benchmark runs.
Investigating worker-executor performance by running executor tests with OTLP tracing enabled and analyzing trace data from Jaeger. Use when diagnosing slow tests, understanding executor call flows, or profiling span durations.
Building or modifying test WASM components in test-components/. Use when a test component needs to be rebuilt, a new test component is needed, or SDK changes require downstream test component rebuilds.
Final checks before submitting a pull request. Use when preparing to create a PR, to ensure formatting, linting, and the correct tests have been run.
Review GitHub pull requests with detailed, multi-perspective code analysis using parallel subagents. Use this skill whenever the user wants to review a PR, asks for code review on a pull request, mentions "review PR", "check this PR", "look at pull request", or references a PR number or GitHub PR URL. Do NOT use for local uncommitted changes — this skill only reviews pull requests on GitHub.