菜单
SkillsMP 已收集 khaneliman/khanelinix 中的 51 个 Skill。打开任一 Skill 可查看来源和详情。
已展示 40 / 51 个已收集 Skill。
GitHub maintainer queues, issue discovery, triage, and creation, pull-request creation and stacking, review authoring and feedback, CI check-fix workflows using gh CLI. Use for read or write work on GitHub issues, PRs, stacks, reviews, or checks.
原文语言：英语
Default lifecycle for routine bug fixes, features, refactors, migrations, dependency updates, configuration, scripts, and code. Routes investigation through handoff. Excludes answer-only, diagnosis-only, review-only, architecture-only, large, cross-cutting,…
原文语言：英语
Select concrete model seats, fallbacks, and quota circuits when provider diversity, a named model, or a route retry matters. Caller owns lifecycle, phase boundaries, and final judgment.
原文语言：英语
Audit retained coding-agent sessions, tools, workers, tokens, and routing. Use for retrospective AI-tool efficacy checks.
原文语言：英语
Git workflows for commit planning, branch cleanup, fixup/autosquash, conflicts, bisect, single-reviewer diff review, and multi-commit stacks. Use for local history or change splitting, repair, review, and validation. Use interrogate for multi-model review.
原文语言：英语
Host-only explicit fan-out overlay for /swarm or 'swarm this'. Never an entry workflow; caller owns lifecycle, integration, and final judgment.
原文语言：英语
AI-tool architecture for instructions, prompts, rules, agents, commands, skills, hooks, and provider context. Use for setup, refactoring, or audits needing clear boundaries, deterministic automation, reproducibility, or lean context. Use software-engineering…
原文语言：英语
Adversarial multi-reviewer synthesis for contested or high-risk plans and changes. Use for multi-model review, independent angles, challenge, stress-test, or blind-spot requests. Do not use for routine single-reviewer checks.
原文语言：英语
Cut AI tells, puffery, promotional language, boilerplate patterns, and canned phrasing from writing. Use when reviewing or rewriting documentation, pull request descriptions, commit messages, or user-facing prose to remove AI writing patterns.
原文语言：英语
Optional persistent file-based planning for multi-phase or long-running work that benefits from recovery across sessions or compaction. Use when requested or when persistence materially improves continuity; do not activate merely due to tool count or existing…
原文语言：英语
Explicit durable control overlay for multiple independently landable work units. Use only when the user names $program-orchestration or directly asks to orchestrate a program with dependencies, leases, grants, receipts, and recovery. Existing workflow skills…
原文语言：英语
Explicit durable control overlay for multiple independently landable work units. Use only when the user names $program-orchestration or directly asks to orchestrate a program with dependencies, leases, grants, receipts, and recovery. Existing workflow skills…
原文语言：英语
Spawn N parallel candidates at the same task, pick a base, graft the strongest parts of the losers into it. Use for /arena, 'arena this', 'throw it in the arena', or when one attempt at a non-trivial artifact would lock in the wrong shape.
原文语言：英语
Methods inside a caller-owned lifecycle: diff sizing, work sequencing, refactoring, debugging, commit stacks, verification, context pressure, recurring corrections, and tool choice.
原文语言：英语
Design and run an auditable playbook when no narrower one fits: scaled rigor, a hypothesis loop, and a logged decision trail. Use for /figure-it-out, 'figure it out', a large migration, or unattended work a human reviews later. Use software-engineering for…
原文语言：英语
Explain a subsystem through direct codebase exploration. Use for how-it-works questions, walkthroughs, caller discovery, placement, or layering decisions. Use why for rationale and software-engineering for repository-wide evaluation.
原文语言：英语
Durable cross-provider user and project memory. Use when asked to remember or persist, when a request depends on prior work, decisions, recurring issues, or preferences, or when substantial research yields a reusable lesson. Provider-native memory is not a…
原文语言：英语
Reconstruct recent working context from local plans, memory, transcripts, and the shared record into a current-state brief. Use for 'recall my work on X', 'catch me up', 'what have I been working on', or 'where did I leave off'.
原文语言：英语
Bounded requirements interview for an explicit interview request or unresolved material product choice. Discover facts first, then ask frontier questions with recommendations. Do not use for routine clarification.
原文语言：英语
Repository or subsystem architecture, maintainability, and large-change planning. Use for architecture-only contracts, invariants, boundaries, failure behavior, evolution, design review, or implementation-ready validation. Use ai-tools-architect for agent…
原文语言：英语
Write or review concise technical prose for docs, procedures, comments, commits, releases, incidents, and postmortems. Preserve facts, caveats, figures, code, commands, links, and tables. Claim full ASD-STE100 conformance only with the official current…
原文语言：英语
Persistent browser and Electron debugging through a harness-native interactive runtime, with Nix-provided Playwright and browsers when available. Use for stateful iterative QA; use playwright for one-shot terminal automation.
原文语言：英语
Use when the task requires automating a real browser from the terminal (navigation, form filling, snapshots, screenshots, data extraction, UI-flow debugging) via the bundled Nix-backed Playwright wrapper script.
原文语言：英语
Mine the session transcript with three review lenses and route durable learnings into skill edits or memory notes. Use when the user says reflect, after a complex task lands cleanly, or after a correction or new workflow worth keeping.
原文语言：英语
Bevy workflows for ECS, migrations, first-party UI/input/rendering/assets/shaders, plugin selection, scenes, scheduling, hot reloading, streamed-world expansion, world authoring, diagnostics, validation, and BRP/MCP control. Use when building, debugging,…
原文语言：英语
Build or iterate on HTML/JavaScript games with a short Playwright verification loop, screenshot and console inspection, and render_game_to_text state checks. Use for playable browser-game implementation or debugging; use frontend-design for non-game…
原文语言：英语
Guide for creating high-quality MCP (Model Context Protocol) servers that enable LLMs to interact with external services through well-designed tools. Use when building MCP servers to integrate external APIs or services, whether in Python (FastMCP) or…
原文语言：英语
Diagnose memory inefficiencies, capture high-resolution memory profiles, and execute architectural refactoring across C++, Rust, TypeScript/Node.js, .NET, and Python. Use when troubleshooting memory leaks, out-of-memory (OOM) errors, or heap fragmentation.
原文语言：英语
Deterministic SARIF reporting and partitioning. Use when inspecting .sarif or .sarif.json static-analysis output, summarizing findings without loading full JSON into chat, or splitting results by rule, path, severity, or balanced chunks.
原文语言：英语
Use only for explicit repository security requests: secure-code review, threat modeling, or security ownership and bus-factor analysis. Do not trigger for generic code review; prefer dedicated security scan skills when installed and requested.
原文语言：英语
Keep a reviewable decision trail for long-running or unattended work: a TSV log with one row per decision (what, why, evidence, result). Local by default; commit it when a reviewer needs the trail to trust the result. Use for /show-me-your-work, autonomous or…
原文语言：英语
Create or update portable agent skills with precise triggers, progressive disclosure, deterministic scripts, references, assets, and optional harness metadata. Use for skill package design or scaffolding; use ai-tools-architect for full AI-configuration…
原文语言：英语
General software bug diagnosis for broken, failing, throwing, flaky, or incorrect behavior. Read-only by default. Inside a requested fix, engineering-workflow owns the lifecycle. Use performance-forensics for latency, CPU, throughput, I/O, contention, or…
原文语言：英语
Read-only measured performance diagnosis for latency, CPU, throughput, I/O, contention, traces, or hillclimbing. Improvement is a Ground or Verify method inside a caller-owned mutation lifecycle. Use memory-profiler for leaks, OOM, or heap fragmentation.
原文语言：英语
Red-green-refactor Implement method inside a caller-owned lifecycle for explicit TDD or test-first requests. Requires red evidence, minimal green implementation, and rerun validation.
原文语言：英语
Create or audit a deterministic project-local verification harness inside a caller-owned Verify phase. Direct use is audit-only. Do not use only to run an existing check.
原文语言：英语
Sketch types, signatures, and module structure before code, then implement against the agreed sketch. Use for /architect, 'architect this', 'design this', or non-trivial features where jumping to code locks in the wrong shape. Use software-engineering for…
原文语言：英语
External research from primary sources with a cited answer. Use for official docs, specifications, APIs, or first-party source. Read-only by default. For repository structure or rationale, use how or why.
原文语言：英语
Find what a change could break somewhere else before it ships, beyond the diff, and prove the one fact it's safe because of by running real code instead of writing it up. Use for 'blast radius of X', 'what could this break', or reviewing a small diff you…
原文语言：英语
TypeScript best practices. Use when reading or editing any .ts or .tsx file.
原文语言：英语