Browse reusable Agent Skills, each with a clear purpose and practical guidance.
Use this skill when you need to validate AI agent skills with dart_skills_lint — running the linter, interpreting failures, fixing violations, and authoring custom rules.
Mandatory checks to run before completing any task that touches md files or dart code in this repository.
Build, sign, and test the legacy jdk15to18 (Java 5) distribution via the build1-5to1-8 script and ant/jdk15+.xml. Use this whenever the user wants to build/sign/run the "build1-5to1-8" version, get "sh build1-5to1-8 test" to pass, reproduce a genuine-Java-5 compile or test failure, or chase a "JCE cannot authenticate the provider BC" / NoSuchMethodError / NoClassDefFoundError that only shows up on a real JRE 5. Captures the build→sign→test ordering, the two different JDKs involved, the machine-only signing step, and the compile-time vs runtime Java-version traps that the modern Gradle build cannot surface.
Guide for running tests in the openai-dotnet repository. Use this when asked to run, debug, or validate tests, or when writing new tests. Explains test modes (Playback, Record, Live), how to identify recorded vs non-recorded tests, environment variable configuration, and what to do when recordings are missing or stale.
Use when modifying hot paths in NUnit — constraint evaluation, equality comparison, value formatting, reflection-based dispatch, collection inspection, or anything that runs once per assertion or per test discovery. Covers allocation patterns, reflection caching, and how the team expects performance claims to be backed up.
Use when writing or modifying tests in NUnit's own test projects, or when making a behavioral change to production code that needs test coverage. Covers test structure, attribute choice, helper visibility, platform guards, and which test projects are real.
Investigating Autumn request logs and Stripe webhook deliveries — debugging unexpected charges, plan-change history, denied feature checks, balance and usage discrepancies, failed API calls, and out-of-sync subscriptions. Use when the user asks why something happened to a customer — a charge, a state change, a denied check, missing or unreset usage — or wants to see a customer's request timeline or log analytics.
Migrate a pytest integration test to the test-loop framework. Use when converting a pytest from pytest/tests/ to a deterministic test-loop test.
Debug and test the tweb (Telegram Web K) MTProto stack from Node.js using its Vitest harness. Use when investigating dialog/message/unread/forum bugs in tweb, when verifying client-side fixes against the real Telegram server, when reproducing race conditions in apiManager / appMessagesManager / dialogsStorage, or when adding new server-verified e2e tests. Covers the harness in src/tests/api/, single-account synthetic tests, and dual-account end-to-end tests.