写作与沟通
2,906 个结果
- Plan Objectivegaurav8936Turn an objective into a visible, editable step plan for One Roof — steps, agent and model per step, dependencies, done-when checks and expected cost — written as Markdown plus a oneroof-plan JSON block.
- PostmanalivirgoBuild Postman collections and environments, write request scripts, and run API smoke checks with Newman.
- PrismicprismicioUse whenever Prismic is mentioned or relevant in any way — including questions, lookups, configuration, repository settings, content modeling, content types, slices, content (pages, assets, releases), localization, previews, API tokens, webhooks, syncing local models, or reading Prismic documentation. Load on any reference to Prismic, even for simple questions.
- Project IndexerxdrshjrGenerate and use a project index for quick codebase understanding in new sessions. Scans the workspace, extracts exported symbols, writes .claude-index/index.md plus config, and injects a Google-style Clean Code Guidelines section into CLAUDE.md. Use when the user says "explore the project", "generate index", "regenerate index", "更新索引", "生成索引", or presses Ctrl+I.
- Prompt Polishhost452bUse when a user hands you a raw question or task and wants it rewritten into stronger bilingual LLM prompts. Triggers - "帮我优化提示词 / 把这个问题改写成更好的 prompt / 提示词重构 / 提示词策略优化 / optimize this prompt / prompt polishing", or any ask to make a question more likely to get a correct, well-formatted answer.
- Prompting Subagentsrhawk117Build, review, or repair the dispatch packet for a mightymodels worker: choose the right worker (code-scout, web-scout, qualitylens, fact-checker, engineer, architect, intern, merge-vader, uncle-bob, gitty-up, wingman, terry-toolchain), fill its role template, resolve its model from ticket.yml, and read the report it returns; also the opening prompt for the next mightymodels session.
- Promptlintrhawk117Use when writing, improving, or reviewing a coding-agent prompt for Claude Code.
- Proposal Writerhongtongzhou-design开题报告与研究方案写作规范。当用户要求写开题报告、研究计划书、基金/课题申请的本子框架时使用。按「选题依据→研究内容→技术路线→可行性→进度安排」组织，产出结构化草稿；引用只用 references.bib 已有键，承诺动词按证据强度分级，草稿供人修改定夺。
- Proverka Oss Licenzijzhuga79Локализация РФ. Проверка лицензий открытого исходного кода в продукте: классификация лицензий по силе copyleft (разрешительные, слабый copyleft, сильный copyleft), совместимость лицензий, обязательства при распространении, квалификация по российскому праву (лицензионный договор — ст. 1235, 1286 ГК; открытая лицензия — ст. 1286.1 ГК; программа для ЭВМ — ст.
- Prune Ticketrhawk117Close out a finished mightymodels unit of work: the close tool refuses while live work remains (unverified tasks, a live whats-broken.md, undecided or unfixed review findings, unpushed commits), compacts the ticket's recorded state into a 30-line archive at .mightymodels/archives/<slug>.md with a JSON sidecar (contracts and results, decisions, review outcomes), marks the ticket closed, proposes…
- Python Code ReviewtasanakornReviews Python code for PEP8 compliance, type hints, modern best practices (Python 3.9+), and code quality. Detects anti-patterns, validates documentation, and provides severity-based recommendations. Integrates with modern tooling (ruff, mypy, uv).
- Release NotesBubblegunnUse when writing release notes, a changelog entry, a status update or a "what's new" message for people who do not read code. Makes the agent lead with what the reader can do now, keep one line per change, explain every term, count every number, and list what is not in this release.
- Repo Mapgaurav8936Write or refresh docs/codebase-map.md — what the repository is, how to run it, its entry points, a one-line-per-folder tour, and the invariants that must not break — so any agent can orient in one read.
- Reportingskrrt-shHow to report to the user, when to ask, and how to finish. Plain short words, defined terms, concrete specifics, done or not done per ask, open questions as one numbered list with recommendations at the end, no mid-run questions, no handing work back.
- Research DocsdkmakerResearch and persist coding documentation using Perplexity AI. Use when user asks about documentation, APIs, frameworks, libraries, best practices, coding patterns, or needs to research technical topics. Always checks existing research first.
- Restart Readykl-autonomousKeep a Claude Code session ready to be thrown away and restarted fresh at any moment without losing anything. Everything it knows lives in short .md files in the repo; it reads only what each step needs; it ends each unit of work by saying whether now is a good moment to restart it; and on request it writes the full prompt for its own replacement.
- Reviewantoine-tenaReview a branch's commits and diff before they become public for good - messages, authorship, forbidden trailers, stray files, secrets, fixups to fold - and tidy the history while it can still be rewritten; with `--repo`, review the whole repo before a release (published commits since the last tag, leftover remote branches, worktrees).
- Roblox Agent Projectm4bwavPut a Roblox game's code on disk so an AI coding agent can edit it as ordinary files under Git: choose between Studio's Script Sync and Rojo, set up the Rokit toolchain (rojo, StyLua, selene, luau-lsp, lune, wally or pesde) in a rokit.toml, scaffold or convert a project (rojo init, rojo syncback), keep one writer per script so Studio, sync and MCP edits never fight, and set up luau-lsp for…
- Session Closetornikebolokadze1-cyberClose the working session properly — write this session's episode, update progress.md ("ახლა" + log) and plan.md, record lessons and proposals, reconcile auto-summarised sessions, run the gates and commit. Use when the owner says the session is ending (e.g. "დავასრულოთ", "დავხუროთ", "ვამთავრებთ"), before /clear or a new terminal, or when the status line shows ↻3.
- SkillsmithMalcolmMcNeelyRules for writing a document an agent reads. Use when creating or editing a skill, a CLAUDE.md or AGENTS.md, a rule file, a subagent brief, or a prompt a script sends. Not for prose a person reads: that is unslop.
- Skillwriterflywheel-productHow to write an AI skill to disk so it is portable to any AI and auto-triggers in Claude — the two-file convention (thin SKILL.md adapter + a named <skill>.md guide that holds all the real content), the adapter template and footers, trigger-rich descriptions with the /slash command, sibling references instead of absolute paths, scripts shipped alongside, the base+overlay and family patterns…
- Spectornikebolokadze1-cyberWrite or revise a feature specification (docs/specs/SPEC-NNNN-slug.md) from the accepted intent, the domain facts and the accepted ADRs, then have it reviewed by the spec-reviewer subagent. Use when the owner wants a feature defined, or before planning any feature.
- Specify Preferred Writing StylealivirgoDefine writing tone, information density, and sentence style with concrete examples.
- Stack ResearchgwpicardRead what has changed upstream for every product the kit's recipes name, and what tools that build apps with AI now produce by default, then write a dated note proposing a change to each recipe and part, or none. For a maintainer keeping the recipe menu current. Proposes only; it changes nothing in the kit. Load it by its path when you want it.