ratacat/ratacats-skills

ratacat のプロフィール

記録された最新のソース活動: 2026年9月23日 07:04

SkillsMP カタログ更新: 2026年10月17日 00:09

収集済み skills: 31

GitHub スター: 9
GitHub フォーク: 0

リポジトリの README

README の元ファイル：README.md

README の元ファイルを読む

README 全文を表示

# Ratacat's Skills Agent skills I use every day, shared. A skill is a folder with a `SKILL.md` — instructions an AI coding agent (Claude Code, Codex, Cursor, OpenCode, …) loads when the task calls for it, plus whatever reference docs and scripts the workflow needs. Install one and your agent picks up the workflow automatically; no prompting ceremony. The collection leans toward evidence-first engineering (bug hunts, reviews, naming), prediction-market research, and a few sharp tools. Each skill's folder has a README explaining what it does and when to reach for it — click through the table below. ## Install Every skill installs two ways. Pick whichever fits your setup: ### Claude Code (plugin marketplace) ```sh /plugin marketplace add ratacat/ratacats-skills /plugin install <name>@ratacats-skills ``` ### Any agent (skills.sh) Works with Claude Code, Codex, Cursor, OpenCode, and [others](https://skills.sh). ```sh npx skills add ratacat/ratacats-skills # choose interactively npx skills add ratacat/ratacats-skills --skill conjecture-cascade # install one skill npx skills add ratacat/ratacats-skills --list # list everything ``` Use the skill `name` from the table below as the `--skill` value (it matches the directory name). ## Skills  ### Developer Tools | Skill | What it does | | --- | --- | | [`clarify`](skills/clarify/) | Turns fuzzy repos, plans, diffs, or issue sets into clearer findings, fixes, names, and acceptance criteria using an evidence-first review chain. | | [`crystallize`](skills/crystallize/) | Finds repeated agent work that can be removed or turned into reusable procedures, with evidence, exception conditions, and a small first change. | | [`ebook-extractor`](skills/ebook-extractor/) | Extract plain text from EPUB, MOBI, and PDF files with local Python tools so agents can search, quote, or analyze book content. | | [`failure-modes`](skills/failure-modes/) | A systematic code-review checklist for finding hidden bugs, stale contracts, weak tests, drift, duplication, and quiet failure paths. | | [`gpt-pro-loop`](skills/gpt-pro-loop/) | Run a local revision loop that packages project context for GPT Pro review, maps critiques into issues, verifies fixes, and assesses the goal. | | [`jev`](skills/jev/) | Builds features on TypeSafe's Jev model — typed Choice/Score/Noul judgments with calibrated probabilities that code branches on directly, no text parsing. | | [`medium-paywall-bypass`](skills/medium-paywall-bypass/) | Fetch readable Medium or Medium-hosted article text through mirror routes so an agent can summarize, compare, or discuss the content. | | [`name-review`](skills/name-review/) | Boundary-first naming review that maps concepts before recommending small, evidence-backed renames for code, plans, schemas, and docs. | | [`no-process-porn`](skills/no-process-porn/) | Detects when agent process is substituting for delivery and redirects the work into the smallest concrete next action. | | [`refine-tests`](skills/refine-tests/) | A methodical test-suite audit that finds tautological, vacuous, weak, brittle, mislabeled, and missing tests, then strengthens them without ever reducing coverage. | | [`writing-claude-skills`](skills/writing-claude-skills/) | TDD-style skill authoring guide that turns repeatable workflows into focused, tested Claude/Codex skills with clear triggers and resources. | | [`x-undocumented-api`](skills/x-undocumented-api/) | Safe guide to X.com's private web API: verifies current GraphQL operations, request headers, cursors, and secret-safe evidence. | ### Prediction Markets | Skill | What it does | | --- | --- | | [`kalshi-prediction-market`](skills/kalshi-prediction-market/) | A Kalshi domain primer for understanding series, events, markets, Yes/No pricing, settlement, order books, and API or WebSocket data. | | [`pm-deep-analysis`](skills/pm-deep-analysis/) | Price-blind prediction-market research: builds evidence-first world reports and calibrated 0-100 resolution estimates without using odds or prices. | | [`pm-market-analysis`](skills/pm-market-analysis/) | Market-aware prediction-market analysis that verifies exact instruments, rules, liquidity, and risks before producing a forecast, report, or no-publish call. | | [`pm-research-methodologies`](skills/pm-research-methodologies/) | General deep-research scaffold that turns a broad subject into named nodes, sourced claims, provenance checks, synthesis, and open questions. | | [`pm-situation-framing`](skills/pm-situation-framing/) | Reusable PMKNB situation frame: separates facts from hypotheses, maps actors, gates, clocks, cruxes, observables, and update rules. | ### Tools | Skill | What it does | | --- | --- | | [`annas-archive-ebooks`](skills/annas-archive-ebooks/) | Finds book records, checks editions and formats, and uses a local Anna's Archive script to download ebooks when a membership key is available. | | [`conjecture-cascade`](skills/conjecture-cascade/) | Directional thinking engine: fires batches of probes through any bounded target — code, claims, plans, or ideas — and closes each probe with evidence or yield. | | [`graph-researcher`](skills/graph-researcher/) | Deep web research that stores sources, findings, relationships, contradictions, and open leads in a SQLite knowledge graph. | | [`lemma`](skills/lemma/) | Breaks a hard question into a persistent graph of small claims, links each to evidence, and has Jev give every claim a 0-1 credence; false and open claims stay on record. | | [`ml-learning-methods`](skills/ml-learning-methods/) | Explains how machine-learning methods work, how they relate, and when to choose them, with a detailed BERT-to-MAE guide and cited research chapters. | | [`prediction-drill`](skills/prediction-drill/) | Vertical two-phase falsification loop for a single bug: rounds of five cheap-to-check predictions drill down to the root cause, then drill back up to a validated fix. | | [`probe-driven-design`](skills/probe-driven-design/) | Use the smallest useful probe to ground consequential decisions before uncertain detail hardens into design. | | [`show-usage`](skills/show-usage/) | Reads live quota windows across AI coding subscriptions and prints percent used and time to reset in a minimal fixed format. | ### Writing | Skill | What it does | | --- | --- | | [`dianalokada`](skills/dianalokada/) | Switches the agent into Diana's sharp, blunt social voice, grounded in a real tweet corpus for posts, replies, roasts, and punchy takes. | | [`julian-writing`](skills/julian-writing/) | Julian Shapiro's Writing Well handbook condensed into a working method: novelty times resonance, hook-first intros, bad first drafts, and rewriting until a reader effortlessly reaches the end. | | [`remove-ai-sins`](skills/remove-ai-sins/) | Idempotent pass that removes seven prose sins (antithesis frame, named feelings, glossed images, inflated register, moral endings, tied-off questions, meta overload) and otherwise leaves the draft alone. | ### Games | Skill | What it does | | --- | --- | | [`duris-game-knowledge`](skills/duris-game-knowledge/) | Gives practical Duris MUD knowledge for navigation, combat, spells, gear, recovery, and parsing live game output without inventing commands. | | [`duris-group-combat`](skills/duris-group-combat/) | Models Duris group fights around roles, assists, caster readiness, adds, flee risk, and survival-first command pacing. | ### Art | Skill | What it does | | --- | --- | | [`ansi-art`](skills/ansi-art/) | Guides terminal-safe CP437/ANSI art from first sketch to validated fixed-width output, with glyph, palette, format, and rendering references. |  ## How this repo works - `skills/<name>/SKILL.md` — the skill itself: frontmatter (name, trigger description, metadata) plus the workflow the agent follows. - `skills/<name>/README.md` — the human-facing page: what it does, when to use it, how to install it. - `skills/<name>/...` — reference docs and scripts the skill uses. - `.claude-plugin/marketplace.json` and the skills table above are **generated** from skill frontmatter by `bun scripts/sync.ts` — edit the skill, run sync, never hand-edit the generated parts. CI fails if they drift. No `plugins/` tree, no symlinks: what you browse is exactly what installs. Contributor gates live in [AGENTS.md](AGENTS.md); one stable handle per skill — directory name, frontmatter `name`, and install name are always identical. ## License [MIT](LICENSE).

リポジトリの README を読み込んでいます…

このリポジトリの skills

ソフトウェア開発者市場調査アナリスト・マーケティングスペシャリストソフトウェア品質保証アナリスト・テスターテクニカルライター

9 件の職業カテゴリ · 97% 分類済み

skill

職業分類

説明

更新

lemma

未分類

Answer a hard question by building a persistent graph of atomic claims, each given a 0-1 credence by Jev from a shared evidence ledger. Keeps false and unexamined claims. Use for research questions, codebase questions, hypothesis testing, or any claim that…

原文の言語: 英語

2026-09-23

crystallize

ソフトウェア開発者

Use when asked to crystallize a workflow, skill, codebase, or agent runs, or to find where agents repeatedly rediscover methods, apply stable rules, or do unnecessary work. Analysis of opportunities to replace repeated agent effort with reusable procedures;…

原文の言語: 英語

2026-09-22

annas-archive-ebooks

市場調査アナリスト・マーケティングスペシャリスト

Find and download ebooks (PDF/EPUB/MOBI) by title/author, or look up published book content. Pair with ebook-extractor to convert downloads to text.

原文の言語: 英語

2026-09-17

remove-ai-sins

テクニカルライター

Idempotent revision pass for literary and nonfiction prose. Invoke with /remove-ai-sins. Removes seven craft sins only — the not-X-but-Y frame, naming an emotion or theme the detail already carries, explaining an image or joke, inflated register, a summing-up…

原文の言語: 英語

2026-09-17

show-usage

ソフトウェア開発者

Show rate-limit and quota usage across the user's AI coding subscriptions (Claude, Codex, Gemini, Kimi, MiniMax, Z.ai, Qwen, Grok) with per-window percent used and time to reset. Use when the user asks about usage, quotas, rate limits, or when subscriptions…

原文の言語: 英語

2026-09-17

x-undocumented-api

ソフトウェア開発者

Manual-only; never auto-load. Use only when explicitly named (x-undocumented-api / x-api-skill).

原文の言語: 英語

2026-09-17

jev

ソフトウェア開発者

Build with Jev, TypeSafe's System One model — fast structured judgments (Choice, Score, Noul) that code consumes directly. Use when replacing fragile parsing or prompt-and-parse LLM steps with typed decisions, or when designing routing, ranking,…

原文の言語: 英語

2026-09-17

ml-learning-methods

データサイエンティスト

Explain, compare, or choose machine-learning objectives and learning regimes, including masked prediction, self-supervision, generative modeling, reinforcement learning, transfer, and classical methods. Use for ML learning-method questions and training-design…

原文の言語: 英語

2026-09-11

julian-writing

テクニカルライター

Use when writing or rewriting nonfiction prose for readers — essays, blog posts, articles, handbooks, newsletters, threads, intros, or any draft that must hold a stranger's attention. Covers choosing the topic, writing a hook-first intro, outlining,…

原文の言語: 英語

2026-09-09

conjecture-cascade

ソフトウェア開発者

Probe a bounded target from many angles at once — audit for hidden bugs/stale state, stress-test a claim or plan, or brainstorm wide. Falsifiable or generative probes through explicit lenses, each closed with evidence.

原文の言語: 英語

2026-09-04

prediction-drill

ソフトウェア品質保証アナリスト・テスター

Use when a bug, regression, or misbehavior has an unknown cause and the user wants iterative hypothesis-driven investigation — invoked by name ("prediction drill", "drill into this bug"), or when asked to falsify hypotheses and converge on a root cause and…

原文の言語: 英語

2026-08-26

gpt-pro-loop

ソフトウェア品質保証アナリスト・テスター

Run a GPT Pro / pro-cli project revision loop: context package, GPT Pro prompt, adversarial review, design map, revision passes, local verification, goal assessments, handoff.

原文の言語: 英語

2026-08-25

no-process-porn

ソフトウェア品質保証アナリスト・テスター

Automatically use whenever a model starts looking at gates, promotion states, smoke tests, instrumentation, telemetry, observability, dashboards, ledgers, manifests, certificates, audits, reports, handoffs, fresh-eyes reviews, verification theater, or…

原文の言語: 英語

2026-08-19

clarify

ソフトウェア開発者

Chained clarity review of a repo, plan, or work items — combines code review, plan refinement, deep-module architecture, DRY/cruft, and naming analysis. Use for clarify/defuzz/audit/harden/fresh-eyes requests.

原文の言語: 英語

2026-08-05

probe-driven-design

その他コンピュータ職

Use when a consequential design, architecture, process, or policy choice depends on unknowns that reasoning alone cannot settle, or when an investigation plan may be over-scoped, prematurely branched, or full of experiments whose results may not change a…

原文の言語: 英語

2026-07-31

pm-market-analysis

財務・投資アナリスト

Analyze one PM event for PMKNB: identify exact Polymarket/Kalshi instruments (conditionId, token IDs, tickers), assess market forces, resolution/oracle risk, liquidity, and counterparty risk; output forecast, report, brief, proposal, or no-publish.

原文の言語: 英語

2026-07-31

ansi-art

特殊効果アーティスト・アニメーター

Create, review, convert, or embed terminal text art: ASCII/ANSI art, CP437, block ASCII, BBS logos, CLI banners, NFOs, SAUCE/XBIN/BIN/ANS output.

原文の言語: 英語

2026-07-09

duris-game-knowledge

ソフトウェア開発者

Duris MUD reference — load for any Duris task: connecting, navigation, combat, spellcasting, zones, geography, equipment, mobs, output parsing, recovery.

原文の言語: 英語

2026-07-09

ebook-extractor

一般事務員

Extract plain text from ebooks (EPUB, MOBI, PDF) for analysis, processing, or reading.

原文の言語: 英語

2026-07-09

graph-researcher

市場調査アナリスト・マーケティングスペシャリスト

Deep web research with Graph-of-Thoughts exploration — parallel branching, relevance scoring, knowledge-graph storage, synthesis into themed reports. Use for researching topics, companies, people, or concepts.

原文の言語: 英語

2026-07-09

medium-paywall-bypass

ソフトウェア開発者

Read paywalled Medium articles in full, including Medium-hosted publications (towardsdatascience.com, betterprogramming.pub, etc.).

原文の言語: 英語

2026-07-09

pm-deep-analysis

市場調査アナリスト・マーケティングスペシャリスト

Price-blind deep research for a Polymarket/prediction-market event or PMKNB situation — fresh or updated analysis, resolution mechanics, mechanism-only briefs, source discovery, 0-100 likelihood independent of market odds.

原文の言語: 英語

2026-07-09

pm-research-methodologies

市場調査アナリスト・マーケティングスペシャリスト

Manual-only; never auto-load. Use when explicitly named, or for the general research-methodology framework when no dedicated methodology skill fits.

原文の言語: 英語

2026-07-09

pm-situation-framing

市場調査アナリスト・マーケティングスペシャリスト

Direct-invocation-only; never auto-load. Use only when explicitly named to create or update a PMKNB situation frame.

原文の言語: 英語

2026-07-09

refine-tests

ソフトウェア品質保証アナリスト・テスター

Use when auditing, scrutinizing, or strengthening an existing test suite — unit, integration, or e2e — because tests pass but bugs still ship, coverage looks high while trust is low, tests were written after the code or by another agent, tests are flaky and…

原文の言語: 英語

2026-07-09

dianalokada

テクニカルライター

Become Diana. Ukrainian software engineer in NYC. KGB accent. Zero filter. Real corpus mode with full main tweets and top-ranked replies. Once invoked there is no going back.

原文の言語: 英語

2026-07-04

duris-group-combat

ソフトウェア開発者

Use when designing or operating Duris group combat with tanks, assists, assumed targets, caster memorization readiness, bash risk, tracking enemies, aggro adds, or command spam risk.

原文の言語: 英語

2026-07-04

failure-modes

ソフトウェア開発者

Use when reviewing a codebase, uncommitted changes, recent agent work, or one subsystem for hidden bugs, flawed assumptions, drift, duplication, weak tests, broken contracts, or other failure modes.

原文の言語: 英語

2026-07-04

kalshi-prediction-market

ソフトウェア開発者

Context and working knowledge for the Kalshi prediction-market domain. Use this skill whenever the user asks about Kalshi markets, tickers, order books, pricing, settlement, or the Kalshi API/WebSocket.

原文の言語: 英語

2026-07-04

name-review

ソフトウェア開発者

Use when reviewing or designing names for plans, beads, work items, architecture docs, codebases, APIs, schemas, modules, states, or domain vocabulary where terminology drift, overloaded concepts, boundary ambiguity, or AI-confusing naming may exist.

原文の言語: 英語

2026-07-04

writing-claude-skills

その他コンピュータ職

Use when user asks to create, write, edit, or test a skill. Also use when documenting reusable techniques, patterns, or workflows for future Claude instances.

原文の言語: 英語

2026-07-04

収集済み skill 31 件中 31 件を表示しています。

AI アシスタントでインストール

このプロンプトをコピーして、使用中の AI アシスタントに送ってください。

Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/ratacat/ratacats-skills.