plimeor
37 verified skills43 total stars
code-plan
Write evidence-backed coding plans for implementation, debugging, refactoring, migrations, design parity work, and long-running agent tasks. Use when defining, clarifying, refining, or validating a development plan, /goal prompt, implementation approach, scope and non-goals, work sequence, acceptance criteria, regression evidence, verification strategy, or stop condition. Near miss: use code-review when judging an existing diff, spec, or already drafted plan rather than drafting or revising a plan. Also use when the user says `design twice` after a plan and wants an APOSD-style second-design pass over the completed plan.
development3
subagent-delegation
Decide whether and how to use authorized sub-agents, then coordinate delegated work while preserving the main agent's context. Use when the user asks for orchestration, parallel agents, delegation, background workers, context isolation, or when another skill needs delegated research, review, implementation, or verification. Owns host-policy checks, delegation packets, non-overlap, report verification, and stop rules. Do not use to bypass tool policy, infer user authorization, or add coordination overhead to simple single-threaded tasks.
tools3
reconsider
Use before finalizing a non-trivial answer, recommendation, review, or decision to reconsider it and raise its quality, especially when shallow reasoning, context inertia, false framing, overconfidence, unfit analogy transfer, or an obvious-but-missed defect could distort the result. Trigger especially before applying external evidence, familiar frameworks, or comparisons to the user's specific request, and when the user asks to reconsider, double-check, take a second look, or sanity-check an answer. Reconsider the draft against its most likely failure mode, and use independent scrutiny only when it is useful and authorized.
development3
code-review
Review concrete code plan drafts, specs, diffs, and implementation shapes. Use for code-review requests, serious code-plan design critique, and judging whether a proposed direction is sound. Prioritize solution direction, premise validity, logic chain, constraints, alternatives, design shape, contracts, tests, local fit, and actionable findings. Near miss: use code-plan to create or revise plans; use code-scope-gate for pre-spec scope shaping.
development3
meta-gpt-prompt-maintenance
Maintain, audit, rewrite, or upgrade prompt artifacts for GPT-series models. Use when adapting existing prompts to current OpenAI GPT prompt guidance; polishing SKILL.md files, AGENTS instructions, system/developer prompts, product assistant prompts, reusable agent workflows, eval prompts, or grader prompts; or reducing legacy process-heavy prompt stacks into clearer outcome-first instructions. Do not use for ordinary prose editing, model-agnostic prompt advice, new skill creation from scratch, or deciding where durable context belongs before a target prompt artifact is chosen.
development2
codex-session-maintenance
Maintain local Codex Desktop/App state when Codex feels slow, session history is heavy, logs or worktrees are large, or the user asks to inspect, archive, clean up, rotate, hand off, reset, or speed up Codex sessions. Use for inspect-first, backup-first local state maintenance: measure hotspots, create handoffs for useful long sessions, archive stale sessions/worktrees, rotate logs, prune clearly dead project paths only after backup and confirmation, and report concrete verification. Do not use for application code performance, model prompt tuning, live conversation compaction, process killing, automation scheduling unless requested, or deleting user data.
tools2
decision-look-before-leap
Use before answering, recommending, reviewing, deciding, or finalizing a non-trivial response when shallow reasoning, context inertia, false framing, overconfidence, unfit analogy transfer, or obvious-but-missed defects could distort the result. Trigger especially before applying external evidence, familiar frameworks, or comparisons to a user's specific request. Apply a look-before-you-leap sanity check and use independent scrutiny only when it is useful and authorized.
development2
writing-blog-illustration
Generate English image-model prompts for blog or article illustrations in a colorful cartoon infographic style. Use when the user asks for an illustration prompt, image prompt, concept diagram prompt, system-architecture visual prompt, comparison visual prompt, or blog/article illustration prompt, including requests to draw, illustrate, create an image prompt, make an illustration, create an infographic, or write an image prompt. This skill returns text prompts, not generated images; if the user asks Codex to directly create or edit a bitmap image, use image generation instead unless they explicitly want a prompt.
development2
url-reader
Read the main content of a specific webpage URL when the user asks to summarize, inspect, cite, reference, or extract facts from that URL, including "read this", "look at this link", "based on this article", or equivalent reference requests. Owns URL retrieval safety for other skills that need webpage body content; not a broad web search tool.
tools2
writing-reader-feedback
Simulate a specified reader persona reading an article section by section and report raw reading-experience feedback, not writing advice. Use when the user asks how a specific audience would experience a draft, whether readers can understand it, or wants reader role-play. Trigger on reader simulation, reader feedback, reader perspective, whether readers can understand a piece, reading experience, or testing how a target reader reacts. For general writing improvement, draft polishing, structural optimization, or revision plans, use writing-blog instead.
development2
code-test-strategy
Decide the smallest useful test strategy for a coding task without letting tests drive production-only complexity. Use when adding tests, choosing unit/integration/E2E level, reviewing mocks, doing TDD, or when an agent is about to add or modify tests. Gates whether to write tests now, defer them, or report a test gap.
development2
writing-blog
Create, diagnose, outline, rewrite, or polish blog posts and articles with reader expectation management, SCQA, Pyramid Principle structure, and concrete revision guidance. Use for writing from notes, optimizing drafts for structure/clarity/readability, polishing notes into publishable form, article outlines, or feedback on whether a draft reads well. Trigger on blog posts, articles, publishing, writing, first drafts, polishing, diagnosis, optimization, structure, outlines, submissions, newsletters, and reader-oriented writing. For raw reader-experience simulation without writing advice, use writing-reader-feedback instead.
testing2
context-engineering
Route durable rules and context to the right layer — task, project, skill, tooling, hooks, MCP, or global — and produce the smallest directly applicable edit. Use for global rules files (~/.claude/CLAUDE.md, global AGENTS.md), repo-local AGENTS.md/CLAUDE.md, task context packs, hook placement (Codex/Claude Code settings.json), collaboration friction diagnosis, and rule-placement decisions.
tools2
code-scope-gate
Pre-work scope triage for coding, debugging, refactoring, automation, planning, and review. Separate the actual request from candidate implementations, delete unrequested surface, classify boundary changes, choose the next narrow action. Use silently for low-risk work; show a visible gate when the user asks for minimal scope or scope risk affects the next action. Use code-plan for full executable plans, code-review for diff/spec critique.
tools2
writing-humanizer
Humanize Chinese or English drafts while preserving the original artifact, meaning, structure, factual claims, and technical boundaries. Use when the user asks to make writing less AI-like, more human, more natural, less generic, more authored, or when a draft feels over-polished, too symmetrical, generic, translated, or detached from the author's real judgment. Do not use this as an AI-word blacklist, synonym-replacement pass, fact-expansion pass, or request to make technical writing casually chatty.
data-ai1
code-standards-gate
Review a concrete spec, diff, or implementation shape against strict code standards. Use when judging public API/CLI contracts, persisted state, schemas/types, generated output, wrapper behavior, helper abstractions, future-facing options, or test coverage before merge. Do not use for broad pre-spec scope shaping; use code-scope-gate for that.
tools1
ops-url-reader
Read the main content of a specific webpage URL when the user asks to summarize, inspect, cite, reference, or extract facts from that URL, including "read this", "look at this link", "based on this article", or equivalent reference requests. Prefer defuddle.md for public webpage main-content extraction, and fall back only when extraction is empty, incomplete, or errors. Do not use for broad web search, URL string edits in code/config, test execution, private or login-gated URLs, or domains with a more specific available tool such as OpenAI docs or GitHub.
tools1
meta-subagent-orchestration
Decide whether and how to use authorized sub-agents, then coordinate delegated work while preserving the main agent's context. Use when the user asks for orchestration, parallel agents, delegation, background workers, context isolation, or when another skill needs delegated research, review, implementation, or verification. Owns host-policy checks, delegation packets, non-overlap, report verification, and stop rules. Do not use to bypass tool policy, infer user authorization, or add coordination overhead to simple single-threaded tasks.
tools1
ops-codex-session-maintenance
Maintain local Codex Desktop/App state when Codex feels slow, session history is heavy, logs or worktrees are large, or the user asks to inspect, archive, clean up, rotate, hand off, reset, or speed up Codex sessions. Use for inspect-first, backup-first local state maintenance: measure hotspots, create handoffs for useful long sessions, archive stale sessions/worktrees, rotate old logs, prune clearly dead project paths only after backup and confirmation, and report concrete verification. Do not use for application code performance, model prompt tuning, live conversation compaction, process killing, automation scheduling unless requested, or deleting user data.
tools1
meta-context-engineering-project
Project-level context engineering. Use when designing, auditing, or shrinking repo-local AGENTS.md/CLAUDE.md, packing task context for a repo, diagnosing project-specific context drift, or deciding whether a rule belongs in task context, project rules, a skill, tooling/checks, MCP/external systems, or the global layer. Trigger on project context, writing AGENTS for this repo, context packs, rule placement, project-convention drift, misfires, and missed triggers. Do not use for global rules files or ordinary coding/debugging tasks. Formerly named context-engineering-project.
tools1
ops-bear
Use Bear App's local `bear` / `bearcli` command-line interface to read, search, create, append, edit, overwrite, tag, pin, archive, trash, restore, open, or manage attachments for Bear notes on this Mac. Use whenever the user mentions Bear App, Bear notes, Bear CLI, `bear`, `bearcli`, `bear help`, the local Bear database, Bear tags, pins, attachments, note capture, updating/retrieving notes from Bear, Bear MCP server setup, or checking this skill against current local CLI behavior. Do not use for Obsidian, generic Markdown files, web articles, or cloud note services; use the relevant local skill or tool instead.
tools1
code-goal-writer
Write complete, measurable Codex Goal contracts for coding work from user intent. Use when defining, clarifying, refining, or validating goals, /goal prompts, durable coding objectives, acceptance criteria, measurable outcomes, stop conditions, validation loops, Figma or design parity, behavior-preserving migrations, refactors, debugging, implementation tasks, or long-running code agent work. Produces intent, scope, non-goals, acceptance results, verification, checkpoints, pause conditions, and stop conditions before planning or execution.
development1
meta-code-standards-calibration
Evaluate `code-standards-gate` against human PR/MR review evidence or explicit conversation-history corrections. Use when the user asks to benchmark, score, validate, calibrate, or improve the review skill; compare human findings with skill findings; diagnose missed review standards; or decide whether evidence belongs in `code-standards-gate`, project rules, tooling, or should stay local. Do not use for ordinary code review; use `code-standards-gate` for reviewing code directly.
tools1
meta-context-engineering-global
Review, prune, or rewrite global rules files such as global AGENTS.md or ~/.claude/CLAUDE.md. Use when deciding whether durable behavior rules belong in the global context layer across tasks, projects, and sessions; when diagnosing repeated collaboration friction; or when producing directly applicable global-rule edits. Exclude project-local AGENTS.md/CLAUDE.md setup, skill routing fixes, task context packing, and tooling enforcement. Formerly named context-engineering-global.
tools1
knowledge-project-docs-maintenance
Maintain project documentation under docs/ as one current source of truth across README, living specs, implementation plans, and decision records. Use when the user asks to organize, update, split, prune, archive, or clean up docs/specs, docs/plan, docs/decisions, README content, docs language or naming policy, ADRs, implemented plans, README-vs-spec boundaries, spec bloat, or post-implementation documentation. Do not use this as a general code-review or implementation planning skill unless the requested output is documentation maintenance.
development1
context-engineering-global
为全局 rules file 做 context engineering。适用于审查或改写 ~/.claude/CLAUDE.md 这类全局上下文文件，判断某条长期规则是否应进入全局层，或优化跨项目、跨任务、跨会话都生效的协作约束。触发词包括：全局提示词、全局上下文、CLAUDE.md 优化、system prompt、global prompt、global rules。
data-ai
blog-feedback
Simulate a specific reader persona going through an article section by section, producing raw reading-experience feedback (not writing advice). This skill outputs what a reader THINKS and FEELS at each section — confusion, expectation shifts, boredom, excitement — rather than suggestions for how to improve the writing. Use this skill (not blog-writing) whenever the user wants to know how a specific audience would experience their article, or asks to simulate/role-play a reader. Trigger on: "模拟读者", "读者反馈", "读者视角", "读者会怎么想", "读者能看懂吗", "阅读体验", "reader feedback", "simulate a reader", "reading experience", or when the user provides a reader persona/background and asks for feedback on a draft. Also trigger when a user says things like "如果一个产品经理读这篇文章", "这篇文章对新手来说怎么样", "帮我测试一下读者的反应". Do NOT trigger for general writing improvement requests like "帮我改这篇文章" or "优化结构" — those belong to blog-writing.
development
ops-url-content-reader
ALWAYS invoke this skill instead of WebFetch when a user's message contains a URL and the task requires reading that page's content. This is the primary URL content reader — it produces cleaner results than WebFetch on virtually all sites. Applies to: summarizing articles, reading blog posts, referencing documentation, checking what a link says, extracting information from any webpage. Trigger whenever the user says things like "summarize this", "read this", "look at this link", "based on this article", "参考", "根据", "帮我 看看", or any variation where a URL's content must be fetched. Do NOT trigger for: writing code that handles URLs, updating URL values in config/.env files, web searches, or running tests. Formerly named defuddle.
development
obsidian-attachment-janitor
Clean up and organize the Attachments folder in the Obsidian vault. Finds and deletes unreferenced attachments, renames referenced ones with a date-prefixed, descriptive naming convention. Use this skill when the user mentions cleaning up attachments, finding unused images, organizing files, renaming attachments, or any variation of "附件清理", "整理附件", "清理图片", "附件重命名".
content-media
defuddle
ALWAYS invoke this skill instead of WebFetch when a user's message contains a URL and the task requires reading that page's content. This is the primary URL content reader — it produces cleaner results than WebFetch on virtually all sites. Applies to: summarizing articles, reading blog posts, referencing documentation, checking what a link says, extracting information from any webpage. Trigger whenever the user says things like "summarize this", "read this", "look at this link", "based on this article", "参考", "根据", "帮我 看看", or any variation where a URL's content must be fetched. Do NOT trigger for: writing code that handles URLs, updating URL values in config/.env files, web searches, or running tests.
development
context-engineering-project
为项目层做 context engineering。适用于 repo-local AGENTS.md/CLAUDE.md 设计、任务级 context packing、项目内上下文漂移，以及判断规则该落在项目规则、skill、tooling 还是当前任务上下文。触发词包括：给这个 repo 写 AGENTS、项目上下文、任务上下文、context pack、模型不按项目约定、误触发、漏触发。
tools
blog-illustration
Generate image generation model prompts for blog post illustrations in a colorful cartoon infographic style with cute character metaphors and pastel color-coded zones. Use this skill whenever the user wants to create a visual or illustration for a blog post or article — system architectures, workflows, concept explanations, comparisons, or any abstract idea that needs a playful visual. Trigger on keywords like "画图", "插图", "图片提示词", "配图", "illustration", "infographic", "image prompt", or when the user shares article content and asks for an accompanying image or diagram to be generated by an image model.
development
meta-project-docs-maintenance
Maintain project documentation under docs/ as one current source of truth across README, living specs, implementation plans, and decision records. Use when the user asks to organize, update, split, prune, archive, or clean up docs/specs, docs/plan, docs/decisions, README content, docs language or naming policy, ADRs, implemented plans, README-vs-spec boundaries, spec bloat, or post-implementation documentation. Do not use this as a general code-review or implementation planning skill unless the requested output is documentation maintenance.
development
code-evaluate-code-standards
Evaluate `code-standards-gate` against human PR/MR review evidence or explicit conversation-history corrections. Use when the user asks to benchmark, score, validate, calibrate, or improve the review skill; compare human findings with skill findings; diagnose missed review standards; or decide whether evidence belongs in `code-standards-gate`, project rules, tooling, or should stay local. Do not use for ordinary code review; use `code-standards-gate` for reviewing code directly.
tools
meta-look-before-leap
Use before answering, recommending, reviewing, deciding, or finalizing a non-trivial response when shallow reasoning, context inertia, false framing, overconfidence, or obvious-but-missed defects could distort the result. Apply a look-before-you-leap sanity check, and prefer independent subagent scrutiny when the conversation context is large or inertia risk is high.
testing
knowledge-obsidian-attachment-janitor
Plan and safely execute cleanup of an Obsidian vault's Attachments folder: identify unreferenced attachments for deletion, rename referenced attachments to date-prefixed descriptive filenames, and verify embeds after confirmed changes. Use when the user mentions attachment cleanup, unused images, Obsidian attachments, renaming attachments, "附件清理", "整理附件", "清理图片", or "附件重命名". Requires an explicit delete/rename plan before any destructive action. Formerly named obsidian-attachment-janitor.
testing
blog-writing
Help create and optimize blog posts, articles, and polished writing with rigorous structure, reader expectation management, and SCQA methodology. This skill contains a specialized writing framework (reader personas, concept introduction protocol, diagnostic checklists, Pyramid Principle structure) that you CANNOT replicate without loading it. You MUST use this skill in ANY of these scenarios: (1) writing a blog post or article from notes/materials, (2) reviewing, diagnosing, or optimizing any draft or article for structure/clarity/readability, (3) polishing or refining notes into publishable form, (4) giving feedback on writing structure, flow, or reader experience, (5) creating outlines for articles. Trigger on keywords: "博客", "文章", "发布", "blog", "写作", "初稿", "打磨", "诊断", "优化文章", "结构", "大纲", "投稿", "公众号", "读者". Also trigger when the user shares a markdown file and asks to improve it, or asks if something "reads well" or "makes sense to readers".
development