write-research
Write a research note: survey the options and evidence behind ONE decision-informing question, and commit to no decision. ALWAYS apply when asked for research,…
Skills
Find an Agent Skill for your task. Read the guide and copy the install command.
Updated 10-30 00:12
Writing & research
14,444 results
write-spec
Write or revise a spec: capture intent as verifiable requirements, never implementation. ALWAYS apply when asked to write a spec, requirements, or acceptance…
sus-research
Research a decision until evidence can carry it. Use when comparing options, evaluating APIs or products, sizing markets, mapping competitors, studying…
autoresearch-loop
Domain-agnostic metric-driven improvement loop, generalizing Karpathy's autoresearch. Use when you want an agent to discover what to measure for a…
grammarly-api-reliability
Analyze a sanitized Grammarly asynchronous job receipt without sending requests, reading document content, or guessing provider SLAs. Use when a Writing Score…
wjs-explaining-tv-slang
Use when the user asks what an American/British TV show or movie line, catchphrase, idiom, or slang means — "What does this American TV show line mean", "What's the meme with XX", "That line in Friends…", "Why do they say that",…
outliner
Tech column outline generation skill. Use when the user provides source material (articles, tweets, video transcripts, etc.) and requests the generation of a writing outline. Generate 3-5 differentiated outline options (prioritizing the story-driven type), and save each option as a separate file for the user to choose from.
exa
Semantic web search using Exa.ai neural search optimized for AI consumption. Use when meaning matters more than keywords -- find academic papers, similar…
writing-plans
Executable plans with granular tasks and file precision. Trigger: When creating implementation plans with 2-5 minute tasks.
mockumentary-characters
Create mockumentary characters with talking head personas, comedic blind spots, and ensemble dynamics. Use when user needs to develop characters for a…
ai-writing-tropes
Detect and eliminate common AI writing tropes from prose. Use when drafting, editing, or reviewing text to avoid the predictable patterns that mark…
field-books-survey
Conduct comprehensive research on any discipline, sort out the discipline classification system and authoritative book lists for each direction, and output a structured classification document and subdirectory skeleton. Trigger words: 调研XX学科, XX学科调研, survey XX学科, 研究XX有哪些方向.
lw-ai-tell-audit
The pre-delivery AI-tell audit for lyrics (歌词 AI 味诊断). Run this on every draft before it ships. The eight enumerable symptoms that make AI-written lyrics…
lw-japanese
The Japanese language layer for lyric writing (Japanese mora and note-mapping). The counting unit is the mora, not the kana or the syllable - the small tsu, the moraic n and the…
lw-tone-check
The tone-melody conflict detector and how to read it (倒字检测器). The one objective, computable criterion this library has - Mandarin dao-zi, where a syllable is…
google-cloud-docs
Use when official Google Cloud documentation is needed.
dialogue
Diagnose flat dialogue, same-voice characters, and lack of subtext. Use when conversations feel wooden, characters sound alike, or dialogue only does one thing…
drafting
Break through blocks and execute first drafts. Use when the outline is done but the draft isn't happening, when writer's block strikes, when the blank page…
story-sense
Diagnose what any story needs regardless of its current state. Use when a writer is stuck, when evaluating story problems, when a narrative feels broken, or…
writing
Use this skill to create high-quality academic papers, literature reviews, grant proposals, clinical reports, and other research and scientific documents…
x-content-archetypes
Called when users audit their own content structure proportions, judge the viral ceiling of a single post, or are puzzled by "I posted an exclamation/opinion post and nobody responded." Key trigger: "content archetype" "which content goes viral" "viral ceiling/viral…
x-four-saves
Used before posting for self-checking "does this content save readers a few steps"; activated when the user asks "is it worth posting / does it have value / should I post this." Not applicable to: already decided to post and only need to check structural completeness (use x-five-piece-checklist), need to classify content archetypes (use x-content-archetypes), need to rewrite the expression (use...
x-three-translations
Call this when what the user has written reads like a product announcement/official update (e.g. "we launched X," "supports Y feature," "works great") and needs to be translated into external language that readers can take away. Also applies when: the user asks "how do I turn announcement-style phrasing into a strong tweet," "why did no one engage with this tweet" (if the reason is that it only talks about oneself and not the reader), "how do I write feature introductions in an appealing way." Trigger…
goalpro
Use when the user wants to write a high-quality Goal Prompt and a Loop Prompt for continuing to evolve after delivery, turning vague, strategic, multi-step, evidence-insufficient, continuous/automated, or easily derailed requests into an executable, verifiable, pausable goal contract with time parameters. Suitable for writing goals, optimizing task prompts, clarifying done/success criteria, deep…