Menu
O SkillsMP coletou 22 skills de lhohan/agentfiles. Abra uma skill para revisar a origem e os detalhes.
Mostrando 22 de 22 skills coletadas.
Build and sharpen a project's domain model. Use when the user wants to pin down domain terminology or a ubiquitous language, or when another skill needs to maintain the domain model.
Idioma do texto original: inglês
Interview the user relentlessly about a plan or design until reaching shared understanding, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, or mentions "grill me".
Idioma do texto original: inglês
A relentless interview to sharpen a plan or design, producing domain documentation as the conversation progresses.
Idioma do texto original: inglês
This skill should be used for detailed guidance on Jujutsu (jj) VCS operations, including committing, pushing, searching history, and working with revisions/revsets. Use when the user asks "how do I use jj?", "translate git to jj", ask to interact with VCS…
Idioma do texto original: inglês
Stress-test a feature by generating edge-case hypotheses and driving TDD for exposed breakage. Use after completing a feature slice.
Idioma do texto original: inglês
Repository-local guidance for agentfiles commit subjects using area prefixes instead of Conventional Commits. Use when committing or the user asks to commit.
Idioma do texto original: inglês
Inspect the codebase, infer behaviour and constraints, then resolve planning decisions that code cannot answer. Use when planning and inspection alone is insufficient to produce an implementation-ready plan.
Idioma do texto original: inglês
Browser automation via Chrome DevTools Protocol (CDP). Navigate pages, evaluate JavaScript, take screenshots, pick elements, dismiss cookie dialogs, and emulate mobile devices — all without Puppeteer or Playwright.
Idioma do texto original: inglês
Use when change intent or decision is worth remembering. Guides Change Intent Records for non-obvious rationale, rejected alternatives, subtle behaviour, or constraints maintainers and agents must preserve.
Idioma do texto original: inglês
Discourage unnecessary setup growth in this repository. Use when changing agent-setup artefacts or the wiring around them: skills, prompts/templates, agents, extensions, packages, or automation.
Idioma do texto original: inglês
Use when user says they are done, want to finish, close a task, end a session, or asks to "land the plane." ALSO use before completing ANY task to run the quality gate workflow (issue sweep, VCS review, verification, CIR check, commit, close).
Idioma do texto original: inglês
Use when running multiple agents or tasks in parallel against a Jujutsu repo with isolated working directories. Use this skill for jj workspace creation, inspection, comparison, cleanup, and stale-workspace recovery.
Idioma do texto original: inglês
Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.
Idioma do texto original: inglês
Use when you have a spec or requirements for a multi-step task, before touching code.
Idioma do texto original: inglês
Use when implementing features by establishing a walking skeleton (primitive whole) and evolving behavior through stable interfaces from the outside in.
Idioma do texto original: inglês
Use when implementing any feature or bugfix, before writing implementation code
Idioma do texto original: inglês
Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always
Idioma do texto original: inglês
Use when reviewing code changes (diffs, PRs, or commits) that need operational verification, project-rule checks, and actionable findings across bugs, security, quality, and compliance.
Idioma do texto original: inglês
USE FOR web search, research, RAG, grounding, browse, find, lookups, fact-checking, documentation, agentic AI. All-in-one, optimized for AI agents. Pre-extracted, token-budgeted web content, deep research, news, images, videos, places, custom ranking.
Idioma do texto original: inglês
Retrieves up-to-date documentation, API references, and code examples from Context7. Use when the user says "use find-docs", "use context7", "use Context7", "context 7", "ctx7", "contact 7", or asks for current docs for a library, framework, SDK, CLI tool, or…
Idioma do texto original: inglês
Use at session start or before the first version control action to detect whether the repository should be operated with Git or Jujutsu (jj).
Idioma do texto original: inglês
Use when adding an extension, plugin, GitHub repo, npm package, or local plugin and you need a structured security review before installing or enabling it.
Idioma do texto original: inglês