Skill: utilities (1/2)
Các Agent Skill thuộc danh mục utilities , kèm gợi ý tham số và mô tả từ bản phát hành.
20 skill trong danh mục này.
| Skill | Chi tiết |
|---|---|
av:advise | Interview-driven advisory skill. Analyzes a prompt or URL (GitHub issue, spec, doc), scouts the codebase, interviews the user one question at a time to reframe the problem into exact requirements and goals, then delivers honest advice: what to do, what to avoid, better alternatives, benefits and trade-offs. Use when the user asks for advice, a second opinion, a sanity check, requirement reframing before committing to a direction, or wants an existing plan or design pressure-tested (grilled) one question at a time. — Tham số: [prompt-or-url] [--html] [--md] [--wiki] [--github] [--agent] [--yagni] |
av:ariadnev | Task router for ariadnev installs. Classifies the task, activates the right installed skills, chains them into the shortest workflow that fits, and spawns installed subagents at defined trigger points to raise output quality. Use at the start of multi-step, multi-domain, or ambiguous work, or when unsure which skill or agent applies. — Tham số: [task to route] |
av:ask | Answer technical and architectural questions with expert analysis. Use for design decisions, best practices evaluation, solution comparison. — Tham số: [technical-question] [--yagni] |
av:autoresearch | Route bounded, goal-directed iteration to the ariadnev skill that owns the desired outcome. |
av:bootstrap | Bootstrap new projects with research, tech stack, design, planning, and implementation. Modes: full (default interactive), auto (explicit autonomous), fast (skip research), parallel (multi-agent). — Tham số: [requirements] [--full|--auto|--fast|--parallel] [--yagni] [--skip-journal] |
av:brainstorm | Turn unclear intent into an accepted outcome and compare viable approaches before delivery. — Tham số: [topic or problem] [--advice] [--html] [--yagni] |
av:code-review | Review code quality with evidence-based rigor. Supports input modes: pending changes, PR number, commit hash, and codebase scan. Focuses on bugs, regressions, maintainability, reliability, and verification gaps. — Tham số: [#PR | COMMIT | --pending | codebase [parallel]] [--yagni] |
av:codex-goal | Guide long-running Codex goal work with a verifiable stop condition. Use when users mention /goal, goal mode, durable objectives, or autonomous multi-turn Codex runs. — Tham số: <objective | goal draft> |
av:coding-level | Set coding experience level for tailored output. Use for adjusting explanation depth, code complexity, and response format to user expertise. — Tham số: [0-5] |
av:context-engineering | Check context usage limits, monitor time remaining, optimize token consumption, debug context failures. Use when asking about context percentage, rate limits, usage warnings, context optimization, agent architectures, memory systems. — Tham số: [topic or question] |
av:cook | Implement features, plans, and fixes with structured workflow. Use for feature development, plan execution, code implementation pipelines. — Tham số: [task|plan-path] [--interactive|--fast|--parallel|--auto|--no-test] [--tdd] [--advice] [--yagni] [--skip-journal] |
av:copywriting | Conversion copywriting formulas, headline templates, email copy patterns, landing page structures, CTA optimization, and writing style extraction. Activate for writing high-converting copy, crafting headlines, email campaigns, landing pages, or applying custom writing styles from assets/writing-styles/ directory. — Tham số: [copy-type] [context] |
av:debug | Debug systematically with root cause analysis before fixes. Use for bugs, test failures, unexpected behavior, performance issues, call stack tracing, multi-layer validation, log analysis, CI/CD failures, database diagnostics, system investigation. — Tham số: [error or issue description] |
av:docs | Analyze a codebase and create, refresh, summarize, or audit project documentation without imposing a fixed docs layout, including authoring and optimizing the root CLAUDE.md/AGENTS.md agent context file. — Tham số: init|update|summarize|agent-context|llms |
av:fable-thinking | Reasoning protocol distilled from Claude Fable 5. Makes any model reason like Fable — evidence-grounded claims, multi-hypothesis diagnosis, concrete simulation, adversarial self-review, calibrated outcome-first delivery. Its never-skipped Floor check catches simple-looking trick questions models answer confidently wrong, and its Constraint Loop mechanically verifies hard output constraints (banned letters, exact counts, strict formats) that models otherwise rubber-stamp. Use for debugging, review, analysis, decisions, constrained writing, or any task where being right matters more than being fast. — Tham số: [task or question to reason through] |
av:fix | Fix bugs, errors, test failures, and CI/CD issues with intelligent routing. Use for type errors, lint issues, log errors, UI bugs, code problems. — Tham số: [issue] --auto|--review|--quick|--parallel [--advice] [--skip-journal] |
av:folder-context | Create a compact subfolder CLAUDE.md and linked AGENTS.md for durable local conventions. Use when a folder needs agent context beyond the project root. — Tham số: <target-folder> |
av:handoff | Create a portable, redacted Markdown continuation contract that a fresh coding agent can consume to resume in-progress work safely. Use when switching sessions, models, or runtimes, or when preserving decisions, verification state, and blockers. — Tham số: [task focus] [--output PATH] [--include-diff] [--include-status] [--force] |
av:interview-docs | Extract a user's vision and decisions into durable project documents through a guided interview. Use for README, ADR, strategy, principles, and structured-doc authoring. — Tham số: <vision | document-path | topic> |
av:journal | Write chronological technical journals for session reflection and change analysis. Journals preserve work history; they do not replace current docs or ADRs. — Tham số: [topic or reflection] |