调研与学习技能

LearnDeck原典のHTML-to-PPT DOM実測converterで、日本語HTMLからnative文字と図形・個別SVG画像を持つPPTXを作る。

中文简介正在准备中。当前显示原文说明。

Adds educational comments to a code file so it becomes a learning resource, asking for the file (with a list of close matches) if none is given. Use when preparing code for learners, onboarding, or teaching.

中文简介正在准备中。当前显示原文说明。

Use when the user asks for deep or academic research — a literature review, PRISMA systematic review or meta-analysis, fact-check, evidence synthesis, APA 7 research report, or Socratic guidance to sharpen a vague research question. Runs a 13-agent pipeline with full, quick, review, lit-review, fact-check, socratic, and systematic-review modes.

中文简介正在准备中。当前显示原文说明。

Generates ideas with named methods from creative practice (SCAMPER, lateral provocations, constraints, pataphysics, and more), routed to the user's situation. Use for open-ended generative or selective requests: "I want to make or start something", "I'm stuck", "inspire me", "make this weirder", "help me pick", or inventing a research question.

中文简介正在准备中。当前显示原文说明。

Runs an autonomous experiment loop on any programming task: you define the goal, metric, and eval command; the agent edits code, measures, and keeps or discards each change until interrupted. Use when optimizing a measurable target (speed, accuracy, size, score) by iteration.

中文简介正在准备中。当前显示原文说明。

Autonomous experiment loop that optimizes any file by a measurable metric. Inspired by Karpathy's autoresearch. The agent edits a target file, runs a fixed evaluation, keeps improvements (git commit), discards failures (git reset), and loops indefinitely. Use when: user wants to optimize code speed, reduce bundle/image size, improve test pass rate, optimize prompts, improve content quality (headlines, copy, CTR), or run any measurable improvement loop. Requires: a target file, an evaluation command that outputs a metric, and a git repo

中文简介正在准备中。当前显示原文说明。

Curate Claude Code's auto-memory into durable project knowledge. Analyze MEMORY.md for patterns, promote proven learnings to CLAUDE.md and .claude/rules/, extract recurring solutions into reusable skills. Use when: (1) reviewing what Claude has learned about your project, (2) graduating a pattern from notes to enforced rules, (3) turning a debugging solution into a skill, (4) checking memory health and capacity

中文简介正在准备中。当前显示原文说明。

Hand a research run over as a file — read `rs_open action=capabilities` for the formats this build writes, choose between `md`, `docx`, a self-contained `html`, the `ris`/`bib`/`enw` library exports and `tex`, pick `draft=watermark` while working or `draft=clean` to hand it over, and read `build/build.json` for what was built from which bytes. Use when a document is finished enough to leave the run, when a citation style or an export has to be chosen, or when a build refuses.

中文简介正在准备中。当前显示原文说明。

Write the document of a research run — plan the question and the outline with rs_plan, write prose and declare its claims together with rs_write, choose the claim strength the evidence actually holds, follow the section structure the run's kind maps to, read the draft back, and build a watermarked draft before the gated clean build. Use when a run's document has to be written, a section drafted, a claim bound to a source, or a build attempted.

中文简介正在准备中。当前显示原文说明。

Work the verification ladder of a research run with rs_verify — read the ladder report before claiming anything is checked, run the two file rungs (r1-record, r2-locate) that need no reader, record the r3-support verdicts and strength verdicts that only a reader can make, tell a miss from an inconclusive, and know what a clean build will still refuse. Use when a document is written and about to be called checked, when a claim's quote does not match its source, or when a gate or build answers with an integrity refusal.

中文简介正在准备中。当前显示原文说明。

Work the review of a research run with rs_review — read the manuscript and record the panel's concerns with the evidence that grounds each one, keep the author's reply and the re-review's finding in their separate columns, run the devil's advocate with the concession threshold the tool enforces, and write the panel report and traceability matrix a response-to-reviewers letter is built from. Use when a document is ready for review, when a concern has to be answered or re-reviewed, or when an attack or a concession is being recorded.

中文简介正在准备中。当前显示原文说明。

Run a research project as a folder on disk — open the run root first, record the user's own instruction verbatim, open the two frame gates and stop for the human to answer them, and keep the stage the ledger says it is. Use when the user asks to start, resume or check a research run, a paper, a review, a brief, a report, a literature review or a protocol.

中文简介正在准备中。当前显示原文说明。

Choose the design and the numbers a research run can honestly claim — write the structured METHOD RECORD (databases, date of access, search string, criteria, screening counts) that r0 validates, pick the claim strength a design can actually carry, record a statistic as data rather than prose, route to the reporting guideline the design owes, and decide what to preregister. Use when a run's question, design, search or statistics have to be settled before the writing, or when a claim needs a strength or a number attached to it.

中文简介正在准备中。当前显示原文说明。

Record the statements a journal requires and a reader checks — authors with their CRediT contributions, ethics approval that names the body and the reference number, how consent was taken, funding with grant numbers, conflicts of interest, data availability with an address or a reason, and the use of generative AI — as a validated record that the submission folder prints and the manuscript never carries. Use when a run is heading for submission, when an author asks who should be listed, when consent, approval or data sharing has to be stated, or when the venue's AI policy comes up.

中文简介正在准备中。当前显示原文说明。

Get literature into a research run when the plugin fetches nothing — fetch the source yourself with web_fetch / web_search or take a file the person already has, READ it with the pdf_* tools or ocr, then ingest record and text together with rs_cite, bind locators, and cite with [@key]. Use when a run needs sources, when a library export has to come in, when a citation key is unknown, or when a claim cannot be checked against a source.

中文简介正在准备中。当前显示原文说明。

When a bug or regression is reported, the FIRST step — before fixing — is to research how and when it was introduced (git archaeology), then record that finding on the tracking issue/PR. Use the moment a bug, error, broken build, or "this used to work" is reported, before writing a fix. Produces a first-bad-commit (or "not a code regression") note you paste onto the issue.

中文简介正在准备中。当前显示原文说明。

Bounded improvement loop: one artifact, fixed eval score, keep an edit only if strictly better. Use for AR_DESIGN/AR_FUNCTION/AR_ACCEPTANCE stages or standalone bounded iteration.

中文简介正在准备中。当前显示原文说明。

Decision transparency, feedback collection, and debate invitations for AID methodology. Active in ALL phases for ALL roles.

中文简介正在准备中。当前显示原文说明。

AID Phase 0 - Research & discovery. Use for validating problem spaces, identifying stakeholders, defining success metrics, deciding whether to proceed.

中文简介正在准备中。当前显示原文说明。

Drives the Phase 4-5 dev pipeline state machine with sub-agent reviews and bounded autoresearch gates. Use when running /pipeline or executing an approved plan.

中文简介正在准备中。当前显示原文说明。

Pre-PRD Research methodology combining Business Analysis, Creative Problem-Solving, Brainstorming, and Innovation Strategy. Use before writing any PRD for problem validation, market research, competitive analysis, stakeholder interviews, ideation, root cause analysis.

中文简介正在准备中。当前显示原文说明。

AID Memory System for session feedback, learning cycles, and skill improvement. Manages state, feedback collection, and improvement analysis. Load for /aid-init, /aid-start, /aid-status, /aid-end, /aid-improve, /aid-memory, /aid-reset.

中文简介正在准备中。当前显示原文说明。

Create, revise, and validate publication-ready academic paper figures and tables. Use for LaTeX tables, related-work comparison tables, result tables, notation/dataset/taxonomy tables, precise source-data-driven experiment plots from CSV/JSON/logs, generated conceptual figures such as system overviews/pipelines/architectures/threat models, captions, artifact specs, source-data traceability, and paper-ready PDF/SVG/PNG/LaTeX exports. Do not use for prose-only paper writing, self-review, reviewer response, rebuttal drafting, or external literature-management workflows.

中文简介正在准备中。当前显示原文说明。

Resolve, validate, and audit manuscript integrity and applicable requirements for academic manuscripts and paper artifacts. Use when Codex needs to determine active paper rules from manuscript state, paper type, venue, submission stage, or workflow mode; lint a LaTeX paper project; assess submission readiness; create or update a paper policy profile; explain why a rule is active; or validate the paper-policy registries. Do not use as the primary skill for prose drafting, peer review, rebuttal writing, citation discovery, or figure/table creation.

中文简介正在准备中。当前显示原文说明。