Research & Search

samraddhashrivastavatech Skill

Fetch and extract readable page content from a URL. Use before trusting or quoting any web hit — turn a snippet into the actual source text.

samraddhashrivastavatech Skill

Universal web search across providers with normalization — the base capability for finding anything on the open web. Use for ordinary lookups, background gathering, and the first query of any research task.

samraddhashrivastavatech Skill

Search source code across providers. Use when you need actual code implementations, not talk about code — snippets, API usage, and available libraries.

samraddhashrivastavatech Skill

Bounded, relevance-guided crawl from seed URLs. Use to map a site or follow a topic across pages without unbounded fetching — frontier kept in check by depth, count, and domain rules.

samraddhashrivastavatech Skill

Autonomous end-to-end research composing all research skills — plan, execute, track, and cancel a budgeted multi-query investigation. Use for real research tasks, not single lookups.

samraddhashrivastavatech Skill

Fact-check verdict vocabulary with evidence — supported, refuted, unverifiable, misleading. Use as the final gate before a claim becomes a stated fact.

samraddhashrivastavatech Skill

News search with freshness — find recent reporting, coverage of current events, and development timelines. Use when the question is about what is happening now or recently.

samraddhashrivastavatech Skill

Research GitHub repositories, code, issues, and releases. Use when the question is about software projects, source code, maintainers, bugs, or version history.

samraddhashrivastavatech Skill

Persist research graph of Question/Claim/Evidence/Source relations. Use to keep a complex research task's findings structured and reusable across sessions.

samraddhashrivastavatech Skill

Expand a topic into query variants — synonyms, related terms, and alternate phrasings that surface content the original wording misses. Use before deep research to widen the net.

samraddhashrivastavatech Skill

Classify a source's authority, type, and freshness. Use to decide how much weight a specific URL deserves before it enters a verdict.

samraddhashrivastavatech Skill

Refine a query with domain, filetype, exact-phrase, and term-exclusion constraints to cut noise. Use when results are broad, stale, or off-topic.

samraddhashrivastavatech Skill

Understand and apply site:/filetype:/intitle:/inurl:/phrase/OR/- operators and route operator-aware searches to the right source kind. Use when precision matters or a plain query returns noise.

samraddhashrivastavatech Skill

Search papers on arXiv/Crossref. Use for scholarly sources, preprints, and citation-grade evidence.

samraddhashrivastavatech Skill

Structured academic literature sweep — cover a topic across sources with per-query caps for a reproducible survey. Use before writing a survey, thesis chapter, or research proposal.

samraddhashrivastavatech Skill

Verify a claim across independent sources with a confidence verdict. Use before asserting any standalone factual claim.

samraddhashrivastavatech Skill

Community sources — Hacker News, issues, forums, and discussions. Use for practitioner sentiment, failure reports, and what actually broke for real users.

samraddhashrivastavatech Skill

Investigate an error message deeply — reproduce a search across official docs, GitHub, and community reports, then synthesize a diagnosis. Use when the user hits an error and "try Google" isn't enough.

samraddhashrivastavatech Skill

Map claim → evidence → source → location. Use to pull the supporting (and disconfirming) passages behind any claim before it is written up or verified.

samraddhashrivastavatech Skill

Site-targeted official/government research — official sources, regulations, public records. Use when authoritative, jurisdiction-specific evidence is needed.

samraddhashrivastavatech Skill

Seek criticisms, limits, and failures — hunt angles, dissenting evidence, and conditions under which a claim breaks. Use to stress-test anything being relied on.

awill1988 Bundle

Index stacked pull requests across an org and write the message that asks for review — merge order, what each PR does, its tracker ticket, and where review has stalled. Renderings: a team-facing digest, a per-PR split for separate threads, and a nudge naming only what has genuinely gone quiet. Use when asked to list or summarise stacked PRs, say a stack is ready for review, chase reviewers, check what is waiting on whom, or draft a PR status update for Slack. Do NOT use for reviewing the contents of a PR, for writing a PR description or commit message, or for a single PR that is not part of a stack.

prince-vince Bundle

Complete Valyu API toolkit for AI agents. Use this skill when asked to perform real-time search across web, academic, medical, transportation, financial sources, content extraction from URLs, AI-powered answers with citations, or comprehensive deep research reports.

prince-vince Bundle

Research and implement modern scroll animations and micro-interactions for web pages. References Dribbble, Jitter, and Mobbin for current design trends. Use when the user wants animations, scroll effects, micro-interactions, hover effects, page transitions, or says "animate", "scroll effects", "make it feel modern", "add motion", "Dribbble-style", or "Jitter animations".

learningmatter-mit Bundle

Act as a reviewer to critically review research plans, manuscripts, or task summaries, pointing out missing baselines, statistical flaws, and weak assumptions.

learningmatter-mit Bundle

Search and retrieve preprint metadata from bioRxiv and medRxiv APIs for biological and medical research.

learningmatter-mit Bundle

Search and retrieve research papers from ArXiv API for scientific research.

learningmatter-mit Bundle

Query and rank synthesis recipes from Materials Project's text-mined literature database with precursors, procedures, and journal references.

learningmatter-mit Skill

Find relevant simulation workflows in the in-house literature database.

learningmatter-mit Skill

Perform iterative, deep, and comprehensive literature research on a specific materials/chemistry topic.

ken3pei Bundle

このスキルは、ユーザーが "/research-design-review" を呼び出したとき、または調査を始める前に、意思決定、調査目的、仮説、前提、証拠ギャップ、調査範囲、検証方法、成功基準を一問ずつ詰めたいときに使用する。プロダクト調査、技術調査、セキュリティ調査、設計改善、障害原因調査、移行計画、ベンダー比較、運用改善などで、いきなり結論や改善案を出す前に調査設計をレビューする。

learningmatter-mit Bundle

Retrieve extensive literature (PubMed) and patent associated with a specific chemical compound via PubChem.

breroz Bundle

Operational support for application test coverage improvement. Use when Codex needs to inspect existing test suites and coverage reports, identify meaningful coverage gaps, add or update tests, improve edge-case and regression coverage, run coverage/test commands, iterate on failures, summarize remaining risk, and optionally commit or push verified test-coverage changes.

everest-an Bundle

Everest Research Algorithm

alexoberneyer Bundle

Download a video, or summarize and review it, from YouTube, X/Twitter, Instagram and the other sites yt-dlp supports. Transcribes locally with mlx-audio when a video has no captions. Use when the user shares a video link and says "download this video", "save the video from this tweet as X", "review this video", "summarize this", "give me the key points", or "is this worth watching?".

alexoberneyer Bundle

Write a message, reply or short write-up that the user sends under their own name, in their voice, from whatever the conversation holds: research, a support investigation, notes or bullet points. Use when the user says "write a message to <name> about this", "draft a reply", "reply to <name>", "turn this into a Teams message", "email <name> about this", "write this up in my voice", or "/draft". Not for text the user wrote or dictated (polish-text) or for proofreading a post (proofread-post).