Las pruebas débiles ralentizan los lanzamientos y ocultan regresiones. Esta skill guía la estrategia de pruebas, TDD, automatización, revisión de cobertura y comprobaciones web con Playwright.

Las decisiones técnicas a menudo necesitan evidencia, no opiniones. Esta skill guía pruebas exploratorias con límite de tiempo que incluyen opciones, resultados de proof of concept, matrices de comparación, riesgos y recomendaciones.

Los proyectos complejos asistidos por IA pueden perder claridad antes de que comience la implementación. Esta skill guía a los equipos a través de especificaciones por fases, gobernanza, planificación, análisis y ejecución de tareas.

Los equipos de Python a menudo necesitan pruebas, cobertura e informes consistentes sin perder la disciplina de TDD. Esta skill proporciona flujos de trabajo, ejemplos y orientación de informes centrados en pytest para Claude, Codex y Claude Code.

Notification integrations can mishandle recipients, credentials, retries, and delivery results. This skill guides WxPusher implementation and troubleshooting using project conventions, documented API behavior, and mock tests.

La pérdida de suscriptores reduce los ingresos recurrentes y oculta problemas del producto. Esta habilidad diseña flujos de cancelación éticos, ofertas de retención segmentadas y programas de recuperación de pagos.

Web research needs explicit queries and consistent search settings. This skill guides XCrawl search requests with location, language, and result limits, then returns upstream responses.

Web pages contain content that is difficult to collect in consistent formats. This skill guides XCrawl requests for page content, structured extraction, and asynchronous result retrieval.

Planning a crawl requires knowing which URLs exist and how broad the target scope is. This skill uses XCrawl Map to discover URLs with explicit filters, limits, and scope settings.

Web pages contain useful information in inconsistent formats. This skill guides XCrawl requests for selected formats and structured extraction while preserving upstream responses.

Bulk website collection requires clear crawl boundaries and asynchronous result tracking. This skill guides XCrawl requests, filtering rules, and retrieval of original API responses.

Saved prospect lists can contain missing template values, incomplete captures, and text that exceeds product limits. This skill checks selected snapshots locally and produces private reports without modifying records or sending messages.

Services can require agent attestations or signed identity claims before accepting a request. This skill guides Link CLI workflows for obtaining proofs and presenting authorized claims.

Educational short videos require research, clear animation, and careful music timing. This skill coordinates agent work and local tools to produce narration-free vertical videos.

Vague prompts can omit primary sources, specific methods, and reliable checks. Deepen rewrites the same task with concrete evidence and validation while preserving its scope.

Markdown articles can lose formatting when pasted into the WeChat editor. This skill guides HTML generation with inline styles and local previews for manual publishing.

Short-video links often separate useful metadata from spoken content. This skill combines TikHub data queries and Qingdou transcription, then saves transcripts as organized Markdown files.

dbskill installations can become outdated across plugin and Skills CLI channels. This skill selects the update workflow, handles retired entries, and reports completion or failure.

Complex questions often receive unsupported theories or misleading historical comparisons. This skill combines theory checks, structural case comparisons, and counterexamples to develop answers with clear conditions.

Audience reactions can be difficult to explain from engagement numbers alone. This skill uses textual evidence and communication theories to explain emotional responses and sharing motives.