Showing 36271–36300 of 77205 skills
Synthesize a feature brief or PRD from existing context and file it in Beads. Use when the user requests a product requirements document.
jupes/agent-forge-harness
Investigate a bug and record its diagnosis and fix plan in Beads. Use for root-cause triage; continue the coding workflow when implementation is already requested.
jupes/agent-forge-harness
Extract or refine a domain glossary from conversation and existing terminology. Use for canonical terms, aliases, and ambiguous domain language.
jupes/agent-forge-harness
Map an unfamiliar file, feature, or module to its callers, dependencies, and system behavior. Use when broader code context is needed.
jupes/agent-forge-harness
Commit everything in logical chunks and push to remote
n0an/VivaDicta
把一句话需求转成可复现、可验收视频工程的乔木智能剪辑导演。Use when the user asks to create, plan, edit, remix, explain, narrate, subtitle, animate, composite, or render a video—including one-line requests such as “制作一个科普视频:介绍 LLM 中的 RL”. Builds English-learning movie mixes, person profiles, explainers, code-drawn motion graphics (UI morph loops, showreels, kinetic-type promos, line-art/pixel/Three.js animation, lyric MVs), product launch videos, cinematic shorts, social clips, course/PPT videos, AI image/video/TTS/music-assisted stories, stock-footage stories, or multi-source montage. Routes across 33台词, ClipSeek/open media, local files, ListenHub, image generation, HTML/SVG/Canvas motion graphics with beat-grid storyboards and determinism probes, Manim, ffmpeg-full, captions, sound design, frame-0 video covers with YouTube thumbnails, and release checks.
joeseesun/qiaomu-cut-skill
Use to claim, release or list OBPI work locks.
tvproductions/gzkit
Use to create or refresh governance constitution artifacts.
tvproductions/gzkit
Use to run Gate 2 for an ADR and record the result.
tvproductions/gzkit
Use to record ADR or OBPI semver renaming migration events.
tvproductions/gzkit
Use to claim, release or list OBPI work locks.
tvproductions/gzkit
Use to decompose an ADR item into an authored OBPI brief.
tvproductions/gzkit
Use to report artifact lineage, relationships or readiness.
tvproductions/gzkit
Explain configuration examples without executing code.
NVIDIA/SkillEvaluator
Track and compare product prices across Amazon, Walmart, Target, and Google Shopping with Decodo — check a current price, find the cheapest seller, or watch a product for price drops over time. A workflow on top of the Decodo CLI, which handles JS rendering, anti-bot, and geo/currency. Reach for this when the user wants to monitor or compare prices, detect a discount or price drop, or build a price-tracking log — instead of scraping retail pages by hand with curl, requests, or Playwright.
Decodo/agent-skills
Track development session events in a daily markdown changelog, including file changes, test results, and key decisions.
claude-world/director-mode-lite
对 VLink 全仓或指定范围执行只读重度代码评审,按功能规划覆盖范围和 subagent,逐文件检查正确性、并发、性能与规范,经对抗复核后给出中文 证据结论。用户要求"全面 review"、"深度检查全仓"、"大规模审计代码" 或"对抗评审"时使用;明确要求落盘时生成报告。
thun-res/vlink
审计 VLink 代码、doc、Doxygen、Python/C API、测试、示例、版本、 CMake、conanfile.py、Android.bp、vcpkg、packup 构建打包面、Agent 索引以及 .github 的 workflows、scripts、Docker、模板、Wiki 与治理 文件同步关系,定位 API、功能或工程入口变化后的遗漏和过时描述。公开 API、功能入口或文档结构变化时必须使用;用户要求"检查文档同步"、 "API 改了查遗漏"、"同步构建打包"、"同步 .github"、"同步审计"或 发版前核对时使用。
thun-res/vlink
Use to drive a guided release pipeline — version bump, changelog drafting, doc review, test/lint, PR creation, tag, GitHub release, and back-merge. Mixes agent steps (the agent does them) with user-only steps (push / merge / tag, blocked by policy and printed for the user). Trigger when the user says "cut a release", "release v0.6", "/dev-release", or asks to bump version + changelog + tag.
joedaws/agent-techniques
Use to run a periodic codebase-health pass — dependency audit, lint, test suite, dead code / TODO scan, doc freshness, task hygiene, and build artifacts. Conservative by default (small fixes inline, larger work files a maintenance task). Trigger when the user says "maintenance", "health check", "audit dependencies", "/maintain", or asks for a periodic project tidy-up.
joedaws/agent-techniques
Use BEFORE implementing code in a project set up with this kit — loads the language rules for detected manifests, the project's CLAUDE.md, and the active task's prior handoff annotations. Replaces the heavy context that would otherwise be re-injected on every prompt. Trigger when the user says "preflight", "load the rules", "ground yourself", or before starting a fresh implementation session.
joedaws/agent-techniques
Use as a pre-commit quality gate — review the working diff for stub patterns (TODO/FIXME/`unimplemented!()`/`todo!()`), debug leftovers (`dbg!`, stray `console.log`/`println!`, commented-out code), then run the project's lint, format, and test suites and verify the active task has a handoff annotation. Prints a PASS/FAIL checklist. Trigger when the user says "review my changes", "/review", "ready to commit?", or asks for a final pass before committing.
joedaws/agent-techniques
Use to start or resume work under the taskwarrior convention — find or create the task for this repo, read prior handoff annotations, log decisions/gotchas during work, and leave a handoff annotation before finishing. Trigger when the user says "start work", "resume work", "log a task", "handoff", "what was I doing", or before beginning a fresh implementation session.
joedaws/agent-techniques
Use when a Task asks you to change an App Work — a running open-source application the owner forked, copied or linked into Ever Works — for example "add an SMS reminder", "fix the failed deployment of 3f9c2ab", or a Goal iteration or Mission Task filed on that App Work. Covers planning within the app's size guidance, respecting its protected paths and human-merge paths, running its declared checks, and writing a reviewable pull request against the branch the app deploys from.
ever-co/ever-works
Use when complex problems require systematic step-by-step reasoning with ability to revise thoughts, branch into alternative approaches, or dynamically adjust scope. Ideal for multi-stage analysis, design planning, problem decomposition, or tasks with initially unclear scope.
DangQuangSE/team-development-skills
根据循环训练日与营养目标推荐麦当劳搭配,比较套餐价格和营养,记录实际摄入与消费,生成月末年末吃麦网页及分享长图。用户要按计划吃麦、顺延训练日、查账本或回顾吃麦与健身目标时使用。
tomsbin/mcd-fit-planner
Ask which skill or flow fits your situation. A router over the skills in this repo.
gentleeduck/duck-mc
Review a diff against a fixed point on two axes - Standards (does it follow what this repo documents?) and Spec (does it do what the ticket asked?) - kept separate so neither masks the other. Carries the Fowler smell baseline plus the findings that have actually recurred in IRYSS audits. Use to review a branch, a PR, work in progress, or "review since X".
gentleeduck/duck-mc
The original v1 taste-skill, preserved for projects depending on its exact behavior. The current default is `design-taste-frontend` (v2 experimental), which is a substantial rewrite. Use this v1 install name only if you need exact backward compatibility.
gentleeduck/duck-mc
Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.
gentleeduck/duck-mc