Browse Skills

Server API Change Verification verifies server and API changes by starting a server, sending curled requests that hit changed code paths, and comparing the full response to expected behavior. Use it when reviewing code changes to confirm new headers, status codes, or rate-limit behavior actually work.

Verify drives the affected runtime surface of a code change to observe behavior end-to-end, providing evidence that the change works before commit. Use it before committing nontrivial changes, not for tests-only or docs-only diffs.

Verify CLI Changes provides an example workflow for verifying a CLI change by building, running the command with relevant arguments, and comparing stdout/stderr and exit codes to expected results. Use it when you need to verify that a CLI modification works as claimed without introducing regressions.

BigCode Evaluation Harness evaluates code generation models on HumanEval, MBPP, MultiPL-E, and other benchmarks with pass@k metrics. Use it to benchmark coding ability, compare models, and measure multi-language code generation quality.

Reproduce an EmDash admin UI bug. Boots a demo with bgproc, drives the admin with agent-browser using the dev-bypass session, and captures the reproduction as screenshots plus a written transcript.

Adversarial code review that assumes bugs exist and hunts for them. Use to audit code for correctness, stress-test pull requests, or when you need a hostile reviewer to find every flaw.

Review a pull request for real bugs, regressions, and convention violations. It enumerates candidate issues, verifies them against the code, and returns structured findings and a verdict — no linter or test execution.

Trace from a reproduced symptom to the source code that causes it, identifying the specific file and approximate line. Rate confidence honestly.

Diagnose traces a reproduced symptom to the source code that causes it. Use it when you need the file, line, and fix direction for a failing test, console error, wrong response, or bad UI.

Write the fix when verify says bug and diagnose says high confidence. Follow EmDash conventions, confirm the reproduce test now passes, run lint and typecheck, stage but do not commit.

Fix implements a diagnosed bug fix after verify confirms a reproducible issue and maintainers trigger the fix flow. It updates the pinned file, proves the repro test passes, and runs lint and typecheck before a candidate is pushed.

Public Site Reproduction verifies bugs in the rendered public site by driving public routes in agent-browser and capturing screenshots, DOM slices, and a replayable transcript. Use it when a bug affects Astro pages, public routing, RSS, sitemap, or rendered content.