jurgendn

@jurgendn source repo

88 skills
0 followers
6 hours ago last updated

▌
Init Workspace · jurgendn bundle
Initialize a structured Obsidian-style workspace for learning, research notebooks, papers, academic applications, talks, test preparation, theory, and similar note-centric knowledge work. Use when the user asks to set up, start, initialize, scaffold, or bootstrap a vault/workspace for work organized primarily through notes and written artifacts. Do not use for software repositories, software applications, experiment codebases, or other coding projects. Ask one short intake batch covering the work and its definition of done, project title plus confirmed target directory, and any differences from the typical case. Then use a ready template as-is, adapt the closest template while keeping its contract internally consistent, or synthesize a lean bespoke vault when none fits. Write only into the confirmed target vault; never modify this skill's templates during a run. This skill creates structure and routing, not the downstream content itself.

0
installs
▌
Gre Issue Essay · jurgendn bundle
Draft, score, and improve GRE Analytical Writing "Analyze an Issue" essays against the 0–6 rubric. Use this skill whenever the user shares a GRE Issue prompt or essay ("score my GRE essay", "AWA feedback", "how do I structure the Issue task", "what does a 5.0 essay need that a 4.0 doesn't"), wants a model essay, asks about the 30-minute timing, or wants targeted practice on position-taking, examples, or counterargument handling. Covers all six official instruction variants (agree/disagree, recommendation, claim-with-reason, two views, consequences). For IELTS essays use ielts-writing-task2; for auditing whether an argument's logic actually holds use argument-audit; for SOP/application essays use apply-sop-writer.

0
installs
▌
Gre Quant Coach · jurgendn bundle
Coach GRE Quantitative Reasoning — quantitative comparison tactics, problem-solving and data-interpretation strategy, content-gap triage, careless-error elimination, and timed drilling. Use this skill whenever the user works on GRE math ("quantitative comparison", "numeric entry", "why did I get this quant question wrong", "I keep making silly mistakes on quant", "GRE geometry/probability/percent review"), wants GRE-style quant practice sets, or asks how to get from a 160s score to a 168–170. Also use for quant pacing and calculator strategy. This is GRE test math (no calculus, no proofs) — for research-level math or ML theory use the theory-* or professor-mentor skills; for the verbal sections use gre-verbal-reasoning / gre-reading-comprehension.

0
installs
▌
Gap Finder · jurgendn bundle
Read a research paper and surface publishable theoretical gaps — unstated assumptions, loose bounds, missing guarantees, unjustified heuristics, and settings the theory does not cover. Trigger whenever the user shares a paper and asks to find weaknesses, gaps, open problems, hidden assumptions, or room to extend it. Also trigger on "lemma autopsy", "weekly referee exercise", or requests to weaken one hypothesis in a lemma and locate the first proof step that breaks before reading the authors' discussion. Do NOT use for literature surveys (literature-triangulation), stress-testing the user's own conjecture (theory-counterexample-hunter), writing a peer review (peer-review-writer), or finding papers.

0
installs
▌
Ielts Grammar Coach · jurgendn bundle
Coach and score English grammar for IELTS, focused on the Grammatical Range & Accuracy (GRA) criterion — estimate a GRA band (X/9) against the descriptors, then diagnose and durably fix the specific errors holding the band down. Use when the user says "score my grammar", "what's my GRA band", "my grammar is stuck at 6", "why is this sentence wrong", "fix the grammar in my essay/answer", "raise my Grammatical Range and Accuracy", "what grammar mistakes do I keep making", or pastes IELTS writing/speaking text and wants the grammar assessed and corrected. Uses coded corrective feedback (mark the error, let the user self-correct), Vietnamese-L1 contrastive diagnosis, a persistent error ledger with spaced re-probes, and targeted drills from the user's own mistakes. Also works in a general-English mode (no band) for everyday text. Distinct from ielts-writing-task2 / ielts-speaking-coach, which score the WHOLE task across all four criteria — this skill isolates GRA and drills the recurring errors. Not for aggregatin

0
installs
▌
Ielts Writing Task1 · jurgendn bundle
Draft, analyse, score, and improve IELTS Academic Writing Task 1 responses (150-word minimum reports on graphs, charts, tables, diagrams, maps, or processes). Use this skill when the user shares a Task 1 prompt, wants a model answer, needs band-score feedback on their own response, or asks how to describe trends, compare data, or structure an overview. Also use when the user asks about Task 1 vocabulary, grammar for describing data, or common Task 1 mistakes. Covers all visual types: line graph, bar chart, pie chart, table, map, process diagram, multiple charts.

0
installs
▌
Ielts Writing Task2 · jurgendn bundle
Draft, analyse, score, and improve IELTS Academic or General Training Writing Task 2 essays (250-word minimum). Use this skill when the user shares a Task 2 prompt, wants a model essay, needs band-score feedback on their own essay, or asks about essay structure, argument development, thesis writing, or common Task 2 mistakes. Covers all Task 2 types: opinion/agree-disagree, discuss-both-views, advantages-disadvantages, problem-solution, two-part questions. Also use when the user asks about Task 2 vocabulary, paraphrasing, or how to avoid common errors. Task 2 carries more weight than Task 1 in the overall Writing band score.

0
installs
▌
Gre Learning Planner · jurgendn bundle
Build a personalised GRE study plan — diagnostic baseline, Verbal/Quant/Writing time allocation, weekly schedule, vocabulary pipeline, official practice-test milestones, and coordination with grad-school application deadlines. Use this skill whenever the user asks how to prepare for the GRE ("GRE study plan", "8 weeks to the GRE, where do I start", "target 325+, currently 310", "how do I balance verbal and quant prep", "when should I take the GRE for fall applications"), wants to know whether/when to retake, or wants their GRE prep integrated with PhD application timelines. Routes actual practice to gre-verbal-reasoning, gre-reading-comprehension, gre-quant-coach, gre-issue-essay, and gre-vocabulary-builder. For IELTS planning use ielts-learning-planner.

0
installs
▌
Gre Verbal Reasoning · jurgendn bundle
Coach GRE Verbal Reasoning Text Completion and Sentence Equivalence — explain items, generate practice sets, diagnose recurring error patterns, and drill the predict-then-match method. Use this skill whenever the user works on GRE fill-in-the-blank questions, shares a Text Completion or Sentence Equivalence item ("why is the answer B", "I picked the trap again"), asks for TC/SE practice questions, wants strategy for two- and three-blank questions, or asks why they keep missing verbal questions despite knowing the words. Also use for GRE verbal timing strategy and section-adaptive pacing. For GRE reading passages and critical-reasoning questions use gre-reading-comprehension; for learning and retaining vocabulary itself use gre-vocabulary-builder; for IELTS (not GRE) use the ielts-* skills.

0
installs
▌
Ielts Speaking Coach · jurgendn bundle
Simulate, score, and give feedback on IELTS Speaking tests (Parts 1, 2, and 3). Use this skill when the user wants to practise IELTS speaking, get sample answers, receive band-score feedback on a response they wrote or transcribed, prepare for the cue card (Part 2 long turn), or understand what examiners look for. Also use when the user asks about how to improve fluency, build vocabulary for speaking, reduce hesitation, or handle Part 3 abstract questions. Covers all four speaking criteria: Fluency and Coherence, Lexical Resource, Grammatical Range and Accuracy, Pronunciation.

0
installs
▌
Apply Cv Builder · jurgendn bundle
Build, revise, and restructure academic CVs for PhD applications, research master's programs, fellowships, research internships, RA roles, and lab applications. Use this skill whenever the user asks to turn a resume into an academic CV, improve research-experience bullets, organize publications/projects/teaching/awards, identify missing evidence for research readiness, or make a CV fit research admissions expectations.

0
installs
▌
Apply Sop Writer · jurgendn bundle
Write, review, score, improve, or iterate on Statements of Purpose and PhD application essays for research programs. Use this skill whenever the user is applying to PhD programs, research internships, fellowships, or graduate programs and asks for SOP strategy, essay structure, research fit, faculty alignment, motivation framing, draft critique, or revision help. Every review and plan checks the statement against the programme's declared theme, specialisation, or track before faculty fit, searching the web for the live official programme page when the target is not already known. This skill is especially relevant for research-focused applications where the essay must connect prior work, future research direction, and program fit.

0
installs
▌
Gre Vocabulary Builder · jurgendn bundle
Build, drill, and retain GRE-level vocabulary the way the test actually uses it — meaning clusters, charge (positive/negative), secondary meanings of common words, and recall drills embedded in Text Completion / Sentence Equivalence contexts. Use this skill whenever the user wants to learn GRE words ("GRE word list", "help me memorize vocab", "quiz me on these words", "what does 'protean' mean and how is it tested"), keeps forgetting words, wants a spaced review routine, or wants practice sentences for new words. Also use when TC/SE practice reveals vocabulary gaps. For solving actual TC/SE questions use gre-verbal-reasoning; for IELTS topic vocabulary use ielts-vocabulary-builder.

0
installs
▌
Ielts Learning Planner · jurgendn bundle
Build a personalised IELTS study plan with weekly schedules, skill-specific drills, and progress milestones. Use this skill when the user wants a study plan, asks how to prepare for IELTS, wants to know how to allocate study time across the four skills, or asks how to improve from their current band to their target band. Also use when the user asks how to use academic paper reading to improve IELTS Reading, wants to integrate reading research papers into their IELTS prep, or asks for a paper-reading routine. Integrates with ielts-reading-strategies, ielts-writing-task1, ielts-writing-task2, ielts-speaking-coach, and ielts-vocabulary-builder.

0
installs
▌
Apply Cv Reviewer · jurgendn bundle
Diagnose an academic CV the way an admissions committee reads it, before any rewriting. Use whenever the user asks to review, assess, critique, or find the weak spots in a CV for a PhD, research master's, fellowship, scholarship, or research internship — including "review my CV", "what will a committee think of this", "are my bullets strong enough", "is my CV too industry", "why does my CV look weak", or "grade my CV like an admissions committee". Decomposes every substantial experience into task, personal contribution, method, and outcome, grades bullets A–D against fixed anchors, and separates intrinsic strength from target relevance. Diagnosis is the deliverable; it rewrites only when the user asks after seeing the diagnosis. Distinct from apply-cv-builder, which builds and rewrites a CV and should be used after this one. Not a scored dossier verdict (apply-dossier-evaluator) and not a pass/fail on a finished artifact against a named bar (professor-critic) — this grades bullets in one CV at any stage.

0
installs
▌
Deck Beamer Academic · jurgendn bundle
Build a compile-ready LaTeX Beamer deck for an academic talk — conference presentation, seminar/colloquium, invited talk, job talk, thesis defense, or a lecture. Use whenever the user wants actual Beamer code for a presentation: "make me Beamer slides", "turn this paper/section into Beamer", "write the .tex for my conference talk", "I need slides for my seminar", "build lecture slides in Beamer", "generate a Beamer deck from these results". Works in two phases — first a per-slide spec, then compile-ready .tex. For deciding the talk's narrative, pacing, and what to cut, use research-talk-planner first (this skill renders that plan into Beamer). For proposals/progress reports in Beamer use deck-beamer-proposal-report; for PowerPoint/Canva business decks use deck-business-report.

0
installs
▌
Deck Business Report · jurgendn bundle
Build a business / industrial report deck in PowerPoint or Canva — executive summaries, stakeholder updates, project/pilot results, quarterly business reviews, R&D-to-business reports, client deliverables, and internal decision decks for a non-academic audience. Use when the user wants PowerPoint or Canva (not LaTeX/Beamer) slides aimed at executives, managers, or clients, organized around situation/recommendation/impact rather than a paper's contribution: "make a PowerPoint report", "build a .pptx for the leadership review", "executive deck for our pilot results", "industrial report slides", "QBR deck", "Canva slides for the client". Two phases — slide-spec, then a real .pptx via the python-pptx builder (or import-ready content + a design spec for Canva). For academic talks/lectures use deck-beamer-academic; for proposal/report decks in LaTeX use deck-beamer-proposal-report.

0
installs
▌
Argument Audit · jurgendn
Interactively audit whether stated evidence or premises establish a claim in a proof, essay, research conclusion, or everyday argument. Use when the user asks "does my evidence support this claim", "is this reasoning valid", "check the logic", "does the conclusion follow", "poke holes in my reasoning", or provides a claim and its support. Proceed Socratically, one question at a time, then converge on a validity verdict plus the smallest fix or counterexample. Use theorem-and-claim-audit instead for a direct report-style audit of formal math or ML claims; citation-auditor to verify whether a source supports an attribution; paper-argument-planner to build a paper's full claim spine; and professor-critic to grade a finished artifact against a named reader.

0
installs
▌
Ielts Progress Reporter · jurgendn bundle
Read marked IELTS Writing feedback in a practice vault and produce a periodic progress report with band trends, recurring errors, revision gains, next priorities, and spaced re-probes of older errors. Use whenever the user asks to "summarise my IELTS progress", "how am I doing", "write a progress report", "track my band scores over time", "what errors keep coming up", "are my essays improving", "review an old error", or requests a monthly/weekly writing report. Also use when the user answers a re-probe recorded in `_dashboard/ielts-reprobe-ledger.yaml`. Read existing feedback; do NOT mark new drafts — use ielts-writing-task1 or ielts-writing-task2. Never invent bands for unmarked attempts and never let recent errors crowd older due errors out of review.

0
installs
▌
Paper To Code · jurgendn bundle
Reproduce a paper's method in minimal, direct PyTorch when no official code is available, to check whether the method works as claimed. Use whenever the user says "reproduce this paper", "implement this paper's method", "there's no code for this paper", "I want to check if this paper actually works", "turn this algorithm/equation into code", or shares a paper/arXiv link and wants a working implementation of its core mechanism. The output is a small, single-file, readable implementation plus staged verification gates (shape checks → overfit tiny batch → baseline comparison at toy scale) and an explicit assumptions ledger for everything the paper left underspecified. Do NOT use for designing training recipes for your own experiments (pytorch-training-recipe / jax-training-recipe), auditing an existing released artifact (reproducibility-audit), or structuring a research repo (research-codebase).

0
installs
▌
Citation Auditor · jurgendn
Audit a paper's bibliography and inline citations for correctness — verifying papers exist, claims match what the cited paper actually says, attribution is accurate, and no citations are hallucinated or misattributed. Use whenever the user wants to check their references before submission, suspects a citation is wrong, wants to verify a specific claim is supported by the cited source, or has received a reviewer comment about incorrect citations. Also use when writing literature-heavy sections and needing to verify that a cited paper actually says what you think it says. This skill catches hallucinated titles/authors, citation telephone (claims distorted through secondary sources), missing citations on factual claims, and wrong-paper citations (citing a follow-up when the original should be cited).

0
installs
▌
Ielts Reading Strategies · jurgendn
Teach, practise, and give feedback on IELTS Academic or General Training Reading strategies. Use this skill when the user wants to improve their reading score, understand a specific question type, practise with a passage, get help with True/False/Not Given or Yes/No/Not Given logic, learn how to match headings, complete summaries, or manage time across 3 sections in 60 minutes. Also use when the user asks why their answer was wrong, how to skim or scan effectively, or how to avoid common reading traps. Covers all 13 IELTS Reading question types and both Academic and General Training formats.

0
installs
▌
Ielts Vocabulary Builder · jurgendn
Build, drill, and apply IELTS-relevant vocabulary for Writing, Speaking, and Reading. Use this skill when the user wants to learn topic-specific vocabulary for a particular IELTS theme (environment, technology, education, health, society, economy, crime, media, etc.), improve word choice in their Writing or Speaking responses, learn academic collocations and word families, practise using less common vocabulary naturally, or understand the difference between informal and academic register. Also use when the user asks for synonyms to avoid repetition, wants to build their Academic Word List knowledge, or needs vocabulary for a specific IELTS topic they are about to write or speak about.

0
installs
▌
Venue Targeting · jurgendn bundle
Select and prioritize publication venues (conferences, workshops, journals) for a research paper. Use whenever the user asks where to submit, which conference fits best, whether a paper is ready for a top venue, how to resubmit after rejection, or when comparing venues like NeurIPS vs. ICLR vs. ICML vs. ACL vs. EMNLP vs. CVPR vs. AAAI vs. ICDM vs. journals. Also use when the user has a contribution in hand and needs to decide between tier levels, theory vs. empirical tracks, workshop vs. main track, or short vs. long paper format. If the paper isn't ready for any venue yet, use submission-readiness-audit first.

0
installs
▌
Gre Reading Comprehension · jurgendn bundle
Coach GRE Reading Comprehension and paragraph-argument (critical reasoning) questions — passage mapping, question-type strategies, wrong-answer elimination, and timed drilling. Use this skill whenever the user works on GRE reading passages or shares an RC question ("why is this inference wrong", "main idea vs primary purpose", "select-in-passage", "select all that apply"), asks how to read dense GRE passages faster, or works on strengthen/weaken/assumption/paradox questions. Also use to generate GRE-style passages and question sets for practice. For fill-in-the-blank Text Completion / Sentence Equivalence use gre-verbal-reasoning; for IELTS Reading use ielts-reading-strategies; for auditing real research arguments use argument-audit.

0
installs
▌
Apply Interview Prep · jurgendn bundle
Prepare for academic selection interviews and conversations — PhD admissions, funded-project or DAAD-style panels, scholarship/embassy panels, MSc admissions, and prospective-supervisor calls. Use when the user says "I have a PhD interview", "DAAD interview next week", "my supervisor wants a call", "what will they ask me", "mock interview me", or has been shortlisted and wants to prepare. Always triage who is across the table first: funder committees, prospective supervisors, and administrative panels assess different things. Distinct from apply-sop-writer (written essays), apply-cold-email-drafter (first contact), and professor-critic (grades the finished rehearsal). If the programme selects on documents only, route out rather than inventing interview prep.

0
installs
▌
Apply Profile Reader · jurgendn bundle
Read an academic-career applicant profile from files, folders, notes, CVs, resumes, transcripts, research statements, project writeups, publication lists, GitHub summaries, recommendation notes, and target-school materials. Use this skill whenever the user asks Claude to read "my profile", "my application folder", "my CV and notes", "my PhD materials", "my academic background", or wants a reusable structured profile before SOP writing, academic CV building, professor emails, faculty/program fit mapping, recommender packets, or application-package audits. This skill is for academic career and research-application contexts; do not use it for generic job-search resumes, industry recruiter profiles, literature surveys, or writing a final SOP/CV directly.

0
installs
▌
Apply Package Auditor · jurgendn bundle
Audit a full PhD or research-application package for declared-theme fit, coherence, missing evidence, narrative consistency, program fit, recommender risk, and deadline priorities. Use this skill whenever the user asks whether their whole application is ready, how to prioritize fixes before deadlines, whether their CV/SOP/research statement/faculty fit/recommenders align, whether the SOP and letters actually answer the programme's declared theme, specialisation, or track, or what weaknesses an admissions committee may notice across materials. Every audit runs the declared-theme screen first, before any academic- or faculty-fit finding, searching the web for the live official programme page whenever the target is not already known from the conversation.

0
installs
▌
Related Work Writer · jurgendn bundle
Write or revise a Related Work section from a source-grounded literature map, annotated bibliography, paper list, or draft. Use this skill whenever the user asks to write related work inside an original research paper, position a paper against prior work, organize citations into themes, explain how their work differs from baselines or prior methods, avoid citation laundry lists, or calibrate novelty claims. This skill assumes the user has sources or a literature map; if they need to discover and verify the literature first, use literature-triangulation before this skill. If the user wants to write a standalone survey, tutorial, review, or perspective paper whose main contribution is the synthesis itself, use survey-paper-writer instead.

0
installs
▌
Survey Paper Writer · jurgendn bundle
Write, plan, or revise standalone survey, tutorial, review, or perspective papers whose main contribution is synthesizing a research area. Use when the user asks to write a survey paper, tutorial paper, review article, perspective article, field overview, taxonomy paper, roadmap, or future-directions paper; needs to turn many papers into a coherent synthesis; wants survey structure, taxonomy design, inclusion/exclusion criteria, visual field maps, tables, tutorial examples, or open-problem sections. If the user only needs a Related Work section inside an original research paper, use related-work-writer instead. If the literature has not been discovered or checked, use literature-triangulation first.

0
installs
▌
Flow Idea To Proof · jurgendn
Orchestrate the path from a raw research idea to a formal claim with a defensible proof design — drawing the hypothesis out of the idea, stating it as a precise theorem/claim, and designing (and auditing) the proof strategy. Use whenever the user says "help me turn this idea into a theorem", "I have a hunch, can we make it rigorous", "design a proof for this", "what would I need to prove this", "formalize this claim", "is this provable and how", or hands over an informal conjecture they want to develop into a formal result. This is a ROUTER that sequences existing singleton skills; it does not write the proof itself. Use flow-learn-new-topic or flow-deep-understanding when the goal is to understand existing work rather than formalize the user's own idea. Use gap-finder when the starting point is an existing paper's weakness rather than a fresh idea.

0
installs
▌
Prompting Claude Models · jurgendn bundle
Design prompts, system prompts, and agent scaffolding for LLM-backed features built on Claude models — LLM-as-judge and evaluation harnesses, RAG over literature or documents, extraction/classification pipelines, and multi-step or subagent agents. Use whenever the user is writing or debugging a prompt for a Claude model: "improve this system prompt", "my LLM judge is inconsistent", "Claude ignores my instructions / over-triggers a tool / is too verbose", "how do I structure this prompt", "why won't Claude use my tool", "tune effort/thinking for this agent", or turning a research workflow into an LLM call. This is prompt *design* craft; for API/SDK mechanics (model IDs, pricing, parameters, tool-use schemas, caching, token counting) use the claude-api reference instead, and for Python code quality use the python-* skills. Model-version-specific behavior changes often, so this skill grounds those claims in the live Anthropic docs rather than reciting them from memory.

0
installs
▌
Results Writeup · jurgendn
Turn verified research findings into clear, calibrated scientific writing — paper sections (Results, Experiments, Discussion), lab notes, technical reports, and grant progress reports. Use this skill whenever the user has run experiments or analysis and wants help writing them up, including casual phrasings like "draft the results section", "write up these numbers", "turn my notebook into a paper section", or "summarize what we found". Also use when the user shares logs, tables, or a notebook and asks for prose. For reviewer-by-reviewer rebuttal strategy, use reviewer-response-strategist instead. The skill's main job is preventing confident-sounding overclaims and hallucinated numbers in scientific text.

0
installs
▌
Research Codebase · jurgendn bundle
Set up and maintain a research-experiment harness so experiments run fast and every paper number traces to one command. Use when starting a new project, running a sweep, aggregating runs into a paper table, or cleaning up a codebase that has become hard to navigate. Trigger on "how should I structure my code", "set up my project", "organize my experiments", "starting a new paper", "run all these configurations", "collect my results into a table", "my code is getting messy", "which run produced this number", or "I inherited this codebase". Built for researchers who are strong in the mathematics and weak in software: it fixes a pipeline contract up front so adding an experiment is one line in a list, never an edit to shared code. Not for designing the training loop itself (pytorch-training-recipe / jax-training-recipe), reproducing one paper's method (paper-to-code), choosing which experiment to run or diagnosing overfitting (experiment-design), or auditing a released artifact (reproducibility-audit).

0
installs
▌
Peer Review Writer · jurgendn bundle
Write a peer review of someone else's paper for a conference or journal — structured summary, evidenced strengths and weaknesses, calibrated scores, and constructive questions. Use whenever the user is the REVIEWER: "I have to review this paper", "help me write a review for NeurIPS/ICML/ACL/a journal", "is this weakness worth rejecting over", "how do I phrase this criticism", "draft my review from these notes", "what score should I give", or they share a manuscript they were assigned. Also use for meta-reviews, emergency reviews, and for serving on a program committee or as area chair — "I'm on the PC for X", "how does bidding/COI work", "how do I manage subreviewers", "I'm an area chair, how do I run the discussion". Do NOT use when the user is the AUTHOR: responding to reviews of their own paper is reviewer-response-strategist, pre-submission self-auditing is submission-readiness-audit, and mining a paper for research openings for their own work is gap-finder.

0
installs
▌
Theory To Toy Cases · jurgendn
Turn abstract math or ML ideas into minimal examples, counterexamples, sanity checks, and tiny simulations. Use for theorem intuition, debugging derivations, constructing stress tests, converting a research idea into small checkable cases, or when a concept is too abstract or a claim needs intuition or falsification.

0
installs
▌
Apply Dossier Evaluator · jurgendn bundle
Score and evaluate a further-education application dossier (PhD, research master's, scholarship/fellowship, or research internship) against a reproducible rubric, and check eligibility for cross-border and scholarship programs. Use this skill whenever the user asks to "rate my profile", "score my application", "how strong is my dossier", "am I competitive for a PhD/scholarship", "evaluate me for VEF/DAAD/Fulbright/Erasmus/MSCA", "does my Vietnamese BSc qualify for an EU PhD", "what's my admit/funding chance", or wants a graded scorecard with per-dimension reasoning rather than a prose readiness report. Distinct from apply-package-auditor, which produces a qualitative coherence audit with no scores; use this skill when the user wants numbers, tiering, eligibility verdicts, or scholarship-mission fit. Use apply-profile-reader first if no structured profile exists yet.

0
installs
▌
Apply Motivation Keeper · jurgendn
Help a PhD student recover motivation, reframe setbacks, and rebuild momentum after rejection, failure, stagnation, or burnout. Use whenever the user feels stuck, discouraged, or lost in their PhD — including phrasings like "I don't know why I'm doing this", "my paper got rejected again", "my experiments keep failing", "I feel behind everyone else", "I'm thinking of quitting", "my advisor doesn't believe in me", "I haven't made progress in weeks", or "is a PhD even worth it". Also use proactively when the user seems demoralized mid-conversation. This is not a therapy replacement — it is structured support for navigating the specific emotional and cognitive patterns that make PhD work hard, grounded in the reality of what research actually is.

0
installs
▌
Industry Problem Framing · jurgendn bundle
Convert vague industry, banking, fintech, enterprise, or operational problems into precise R&D problem statements, stakeholder maps, hypotheses, constraints, and solution search directions. Use whenever the user says a problem is vague, ambiguous, unclear, broad, political, business-driven, or not yet researchable. This skill should usually run before publication-grounded-solution-design when the problem is not specific enough to search literature or draft methods.

0
installs
▌
Flow Learn New Topic · jurgendn
Orchestrate breadth-first onboarding into an unfamiliar topic, method, subfield, or research area — going from near-zero to a working, usable map. Use when the user says "I want to learn X", "help me get up to speed on X", "I'm new to X, where do I start", "give me a learning path for X", "I need to understand this field before I can read its papers", or gives a topic they have no grounding in. Routes through singleton skills with graded exercise checkpoints. Also carries a DISCUSSION MODE in which the user learns by explaining to a cast — an honest novice, an opinionated lab mate, and a professor — instead of taking exercise checkpoints; trigger it on "I learn better by discussing", "question me while I explain", "let's discuss this instead of you lecturing me", "set up a study group", or "run this like a lab meeting". Use flow-deep-understanding when the user already knows the basics and wants mastery of ONE specific thing; use flow-idea-to-proof to formalize the user's own research idea.

0
installs
▌
Flow Paper Lifecycle · jurgendn bundle
Orchestrate a research paper end-to-end — from raw idea through literature grounding, experiment design, implementation, results, writing, submission, rebuttal, and artifact release. Use whenever the user wants to drive a whole paper rather than one stage: "help me write this paper from scratch", "where am I in the paper process", "what should I do next on this project", "take this from idea to submission", "I have results, get me to a draft", "manage my paper workflow", or hands over a project and asks for the path to a venue. Also use it for the writing phase itself — "draft the introduction", "outline this paper", "how should I structure this", "turn this spine into a draft", "my first paper" — which it owns as a phase-gated stage-group (stage 10) rather than delegating. This is otherwise a ROUTER that sequences existing singleton skills with stage gates; it does not do the ideation or experiments itself. For a single narrow section, invoke that section's skill directly (e.g. abstract-and-intro-writer, met

0
installs
▌
Flow Phd Application · jurgendn
Orchestrate a complete PhD / research-program application package end-to-end — from reading the applicant's materials through CV, program & faculty fit, SOP, recommender strategy, professor outreach, a final package audit, and interview preparation once shortlists arrive. Use whenever the user wants to drive the whole application rather than one piece: "help me with my PhD applications", "where do I start with my application", "manage my application process", "I'm applying to PhD programs, what's the plan", "get my materials ready", or hands over a folder of CV/transcripts/notes and asks for the path to submission. This is a ROUTER that sequences existing singleton skills with stage gates; it does not write the SOP, CV, or emails itself. For a single piece, invoke that skill directly (e.g. apply-sop-writer, apply-cv-builder). For driving a research paper use flow-paper-lifecycle.

0
installs
▌
Apply Cold Email Drafter · jurgendn bundle
Draft, critique, and revise concise cold emails to professors for PhD applications, research internships, RA opportunities, lab rotations, or prospective-supervisor outreach. Use this skill whenever the user wants to contact a professor, ask about PhD openings, request a meeting, introduce their research background, mention a paper, follow up politely, or convert professor-fit notes into an email. This skill should trigger even if the user says "email a PI", "message a potential advisor", "reach out to faculty", or "cold mail" rather than "cold email".

0
installs
▌
Apply Program Fit Mapper · jurgendn bundle
Map an applicant's research interests to programmes, tracks, faculty, labs, and departments, and check whether their profile actually matches a programme's declared theme, specialisation, or application focus before writing target-specific materials. Covers PhD, research master's, and joint/consortium programmes such as Erasmus Mundus. Use whenever the user asks which professors or programmes fit their interests, how to shortlist targets, how to compare faculty fit, "does my profile match this programme's focus", "is my background relevant to this specialisation/track/theme", "should I even apply to this one", how to turn interests into searchable research themes, or how to prepare target-specific fit notes for SOPs.

0
installs
▌
Naive Student · jurgendn bundle
Become an honest novice while the USER teaches a concept, theorem, method, or mechanism. Use when the user says "let me explain this", "test me by making me teach it", "pretend you know nothing", "find gaps in my explanation", "repay this knowledge debt by teaching", or wants a Feynman-style teach-back whose misunderstandings expose missing dependencies. Ask naive but structurally pointed questions grounded only in what the user actually said; never invent random confusion and never fake comprehension to be polite. End with a fallible "what I think you said" playback that localizes gaps. Do NOT teach the concept, solve the problem, grade a finished artifact, or run a general quiz.

0
installs
▌
Deck Beamer Proposal Report · jurgendn bundle
Build a compile-ready LaTeX Beamer deck for a research proposal or a research report — grant/PhD/project proposal defenses, thesis-proposal talks, progress/milestone reviews, mid-term/final project reports, and committee or funder updates. Use when the user wants Beamer slides organized around objectives, work packages, timeline/Gantt, milestones, deliverables, budget, risks, and progress-vs-plan rather than a single paper's contribution: "Beamer slides for my proposal defense", "progress report deck in LaTeX", "milestone review slides", "PhD proposal presentation", "project final report in Beamer". Two phases: slide-spec, then .tex. For a single paper's results talk use deck-beamer-academic; for PowerPoint/Canva executive reports use deck-business-report; for narrative/pacing of a talk use research-talk-planner.

0
installs
▌
Jax Training Recipe · jurgendn bundle
Produce an implementable JAX, Flax, and Optax training recipe with explicit PRNG management, pure functions, jit/vmap/pmap usage, gradient computation, Flax state handling, checkpointing, and debugging checks. Use when implementing a model in JAX/Flax, porting from PyTorch, debugging tracing/shape/NaN issues, or designing research code that uses JAX transformations.

0
installs
▌
Theory Proof Sketcher · jurgendn
Convert an informal mathematical idea into a precise theorem, lemma, or proof scaffold with explicit assumptions, dependency structure, and clearly marked gaps. Use when developing a new proof, formalizing intuition, planning a theorem structure, or evaluating whether an argument is actually complete.

0
installs
▌
Banking Use Case Discovery · jurgendn bundle
Discover and prioritize banking, fintech, and financial-services AI/ML/R&D use cases from vague business goals, stakeholder pain points, data assets, or operational bottlenecks. Use whenever the user works on banking problems such as credit risk, fraud, AML, collections, customer intelligence, personalization, document AI, compliance, branch/contact-center operations, treasury, or productivity and needs a structured map of feasible, high-impact, research-grounded opportunities.

0
installs
▌
Gap Motivation Builder · jurgendn bundle
Interactively build the motivation case for filling a research gap — turn "no one has done X" into who is concretely hurt, why the obvious extension doesn't work, why now, and what any solution must satisfy. Use when the user has found a gap (from gap-finder or anywhere else) and asks "why does this gap matter", "is this gap worth filling", "who cares about this", "motivate this problem", "justify working on this", "help me build the motivation", or worries about proposing a method from nowhere. Proceeds Socratically, one question at a time, and converges on a motivation verdict plus solution desiderata — it derives the requirements a solution must meet but never proposes the method itself. Use gap-finder to FIND gaps in a paper; research-idea-stress-test to attack an already-chosen idea or method; literature-triangulation to check whether the gap is already closed; abstract-and-intro-writer to turn a finished motivation into intro prose.

0
installs
▌
Reproducibility Audit · jurgendn bundle
Audit whether a result can actually be reproduced from the available code, data, configs, and documentation. Use before trusting a benchmark claim or releasing your own work. Use for replication, code release checks, paper artifact sanity checks, lab handoffs, catching missing seeds/configs/data before results are cited, or whenever "works on my machine" is a risk. If the user wants to plan what to include in a public artifact package, use artifact-release-packager instead; this skill's job is to test whether reproduction is possible and where it breaks.

0
installs
▌
Pilot And Evaluation Design · jurgendn bundle
Design practical pilots and evaluation protocols for industry R&D, AI, ML, analytics, LLM, banking, fintech, or enterprise solutions. Use when the user has a proposed method or solution direction and needs to validate it with offline tests, backtests, human evaluation, A/B tests, baselines, metrics, ablations, guardrails, and a go/no-go decision. Especially use after publication-grounded-solution-design or industry-problem-framing.

0
installs
▌
Whiteboard Peer · jurgendn bundle
Co-solve the unfinished middle of a research idea, conjecture, proof attempt, model design, or technical mechanism as an opinionated peer at a whiteboard. Use when the user says "think this through with me", "I have a half-formed idea", "argue with this intuition", "work on this proof direction without solving it for me", "be my research peer", or has partial reasoning that needs productive disagreement before a finished artifact exists. Contribute one concrete partial attempt, commit to a falsifiable position, expose its weak point, and challenge at least one premise each round. Never slide into full-solution assistant mode. Do NOT grade finished work, deliver a complete proof/solution, teach a novice, or manufacture disagreement unsupported by the object.

0
installs
▌
Literature Triangulation · jurgendn bundle
Build a source-grounded map of a research topic by comparing papers, benchmarks, code, and genuine disagreements. Use whenever the user wants to understand a research area, not just find papers — including phrasings like "what's the current state of X", "what should I read on Y", "is X still SOTA", "are there good baselines for Z", "what does the field believe about W", "survey this area for me", "what are the key papers on X", "compare these two approaches", or "is this claim actually well-supported". Also use when the user is about to build on a method and should first verify the evidence base is solid. If the user already has a source map and wants polished Related Work prose, use related-work-writer instead. The skill's main job is preventing hallucinated consensus — confidently synthesized positions that aren't grounded in the actual sources.

0
installs
▌
Experiment Design · jurgendn bundle
Design a minimal, decision-relevant experiment with fair baselines, explicit controls, and measurable success criteria — and diagnose a training run that has already gone wrong. Use before large-scale implementation or expensive training (benchmarking, hyperparameter sweeps, large-model training, or whenever a research claim requires rigorous empirical validation), and also when a run misbehaves — "my loss diverged", "training collapsed", "the model won't overfit a tiny batch", "these gains look too good", "is this data leakage", "which hyperparameter should I try next", "how do I summarize these runs into a decision", or "is more data/compute still buying anything" (fit that with the bundled scripts/fit_scaling_law.py rather than eyeballing a log-log plot).

0
installs
▌
Theorem And Claim Audit · jurgendn
Pressure-test mathematical arguments, derivations, and ML paper claims, and deliver a written report-style audit. Use for theorem sketches, proof outlines, step-by-step derivations, ML paper claims, reviewer-mode critique, hidden assumptions, skipped proof steps, invalid generalization, weak empirical support, overclaimed conclusions, or whenever a result looks cleaner or stronger than expected — including "check my derivation", "which step breaks", "did I drop a constant", "does this algebra hold", or "audit this proof". Use argument-audit instead when the user wants to be walked Socratically to the weak link rather than handed a report, or when the claim is an essay/application/everyday argument rather than a formal or empirical one.

0
installs
▌
Flow Deep Understanding · jurgendn
Orchestrate depth-first mastery of ONE specific object the user already partly knows — a paper, method, algorithm, theorem, derivation, or result — until they can re-derive it, criticize it, and find where it breaks. Use when the user says "I want to really understand X", "help me understand this paper deeply", "I get the gist but not the mechanism", "walk me through this method", "defend/criticize this", or "tear this apart so I get it". Routes through singleton skills and verifies gates with graded exercises. Also supports DISCUSSION MODE, where the user masters the object by explaining it to an honest novice, an opinionated lab mate, and a professor; trigger it on "I learn better by discussing", "question me while I explain", "let's discuss instead of lecturing", "run this like a lab meeting", or "grill me on this paper". Use flow-learn-new-topic for breadth-first onboarding into an area; use flow-idea-to-proof to formalize the user's own idea.

0
installs
▌
Banking AI Literature Mapper · jurgendn bundle
Map publications, prior successful cases, datasets, benchmarks, and method families for banking and fintech AI problems. Use when the user needs publication-grounded evidence for credit risk, fraud, AML, collections, customer intelligence, personalization, banking document AI, compliance, contact-center automation, or financial-services R&D proposals. This skill specializes literature-triangulation for banking domain transfer and should feed publication-grounded-solution-design.

0
installs
▌
Professor Critic · jurgendn bundle
Critique a FINISHED artifact the way the specific person who will judge it actually would — a strict-professor teardown with a verdict, not encouragement. Use when the user has something done and wants to know if it survives contact with its real reader: "tear this apart", "would a reviewer reject this", "be brutal, don't be nice", "how will my advisor read this proof step", "will a professor reply to this cold email", "grade this like an admissions committee", "poke holes in my abstract". Requires a named reader and an acceptance bar; refuses to grade blind. Distinct from peer-review-writer (there the user is the REVIEWER of someone ELSE's paper; here it is the user's OWN artifact); from submission-readiness-audit (a paper-only, whole-paper author checklist, not a named-reader verdict on any artifact); and from knowledge-debt-audit (which probes the user's UNDERSTANDING, not the artifact). Do NOT use on mid-draft or thinking-stage work — a REJECT verdict on something unfinished only demoralizes.

0
installs
▌
Figure Table Planner · jurgendn bundle
Plan, audit, or revise the figures, tables, captions, and visual evidence for a research paper. Use this skill whenever the user asks what figures or tables they need, whether a result should be a plot or table, how to present experiments, how to design ablation tables, how to make captions self-contained, what belongs in the appendix, or whether the current visual evidence supports the paper's claims. This skill is especially useful after experiment design and before writing the Results section.

0
installs
▌
Method Section Writer · jurgendn
Draft or revise the method (approach / model / framework) section of a research paper. Use whenever the user needs to write or improve this section — including phrasings like "write my method section", "how do I describe my model", "my method section is too long / too vague / too dense", "how much detail do I need", "how do I present the algorithm", "write the approach section", or "explain my method clearly". Also use when the notation is inconsistent, the section jumps between abstraction levels, or reviewers complained the method is hard to follow. If the paper's overall argument is not yet stable, use paper-argument-planner first. If the abstract and introduction are not yet written, use abstract-and-intro-writer first.

0
installs
▌
Research Talk Planner · jurgendn
Plan and structure a research talk (conference presentation, seminar, invited talk, or job talk) from a paper, set of results, or research overview. Use whenever the user needs to turn a paper into slides, structure a 15/20/30/45/60-minute talk, decide what to cut for a short slot, plan a job talk narrative, or figure out how to present technical content to a mixed audience. Also use when the user asks "how do I present this paper", "what should my slides cover", "how long should each section be", or "I only have 15 minutes". A talk is not a paper read aloud — the narrative arc, pacing, and depth of detail must be rebuilt from scratch for a listening audience.

0
installs
▌
Research Idea Stress Test · jurgendn bundle
Stress-test a research idea before major investment. Use to expose hidden assumptions, weak novelty, confounders, and the cheapest decisive experiment. Use before starting a project, writing a proposal, scaling experiments, or committing major compute/resources.

0
installs
▌
Pytorch Training Recipe · jurgendn bundle
Produce an implementable PyTorch or PyTorch Lightning training recipe with dataset handling, optimization, logging, checkpointing, and debugging checks. Use when turning a paper idea into runnable training code — including new experiments, reproductions, fine-tuning setups, scaling recipes, and debugging unstable training.

0
installs
▌
Apply Program Theme Extractor · jurgendn
Determine what a programme, track, specialisation pathway, doctoral school, or funding call actually selects for — its declared theme, or, where none is published, its operative theme inferred from official evidence. Use whenever the user asks "what is this programme's theme", "what does this call actually want", "is this track themed or open", "the landing page doesn't say what they focus on", "what do these 12 programmes each specialise in", or wants a candidate list of targets characterised before any of them is compared against a profile. Works applicant-free and over many targets at once, producing a reusable theme dossier per target. Distinct from apply-program-fit-mapper, which needs the applicant's evidence and issues the four-way fit verdict — route there once a profile enters. Distinct from apply-research-direction-mapper, which maps one person's research direction rather than an institution's selection scope.

0
installs
▌
Paper Argument Planner · jurgendn bundle
Plan the argument of a research paper from an idea, partial draft, experiment results, notes, or reviewer-facing story. Use this skill whenever the user asks how to structure a paper, clarify the contribution, build a paper outline, connect claims to evidence, decide what belongs in each section, turn results into a coherent narrative, or diagnose why a paper feels unfocused. This skill should trigger before drafting full prose when the paper's thesis, claim hierarchy, or section logic is not yet stable.

0
installs
▌
Artifact Release Packager · jurgendn bundle
Plan and audit research artifact releases for papers, including code, data, models, configs, checkpoints, scripts, environment files, licenses, and artifact evaluation instructions. Use this skill whenever the user wants to release a paper codebase, prepare an artifact evaluation package, map paper tables to reproduction commands, write release instructions, check whether a repository supports the paper's claims, or package models/data safely. Use reproducibility-audit to test whether results can be reproduced; use this skill to decide what to include and how to present the release.

0
installs
▌
Cross Domain Analogy Finder · jurgendn bundle
Find a concept's structural twin in a distant field — to borrow a solution, or to check whether the idea already exists under another name. Use when stuck on a problem and wanting to widen the search ("has anyone solved this shape of problem", "find precedent in another field", "what field has this same structure", "I'm stuck, where else does this appear"), or when checking novelty before investing ("is my idea actually new", "has this been solved elsewhere", "am I reinventing something"). Two modes — OFFENSE borrows a portable solution from a far domain; DEFENSE checks whether your idea already exists. This is high-variance foraging: use it when genuinely stuck or vetting novelty, NOT as a daily driver. Grounds prior-art claims through literature-triangulation; hands the residual mismatch to gap-finder. Do NOT use for finding papers or precedent inside your own field or surveying an area (literature-triangulation), or stress-testing one chosen idea (research-idea-stress-test).

0
installs
▌
Submission Readiness Audit · jurgendn bundle
Audit a research paper, draft, appendix, or submission package before deadline. Use this skill whenever the user asks if a paper is ready to submit, wants a pre-submission checklist, needs to find reviewer-obvious weaknesses, check claim/evidence consistency, verify figures and tables are referenced, inspect limitations/ethics/reproducibility statements, or prepare camera-ready materials. This is the final gate after argument planning, experiment analysis, and results writing.

0
installs
▌
Theory Assumption Extractor · jurgendn
Extract explicit and hidden assumptions from a proof, model, or empirical argument. Use for theorem reading, paper review, proposal critique, debugging proofs, conclusions that appear stronger than the stated setup, or determining which premises are logically necessary, proof conveniences, and where the argument breaks.

0
installs
▌
Theory Formalism Translator · jurgendn
Translate an idea across optimization, probability, information theory, dynamical systems, graph theory, and statistics language. Use for interdisciplinary theory work, reframing a problem into a more tractable formalism, connecting literatures, discovering alternative analytical tools, or determining whether different fields express the same structure under different notation.

0
installs
▌
Apply Research Direction Mapper · jurgendn bundle
Define a professor's recent and trending research direction from publications, lab pages, grants, talks, students, and public materials before writing PhD application fit notes or cold emails. Use this skill whenever the user asks what a professor is working on now, whether a faculty member's research is shifting, how to infer current interests from recent papers, how to prepare for contacting a professor, or how to turn scattered professor evidence into application-ready research-fit angles.

0
installs
▌
Knowledge Debt Audit · jurgendn bundle
Detect when AI assistance has produced output you cannot regenerate yourself, call the loan before it compounds, and re-probe repaid debt on a spaced, variable-cue schedule. "Get the work done" is not "I understand it." Trigger when the user is about to BUILD ON an AI-produced result ("before I extend this", "am I understanding this or just getting it done", "check whether I really get this"), wants a standing audit ("what am I in debt for", "what could I not rebuild alone"), or starts a session or periodic review in a workspace with `debt-ledger.yaml`. Two modes — REACTIVE (circuit breaker at the borrow moment) and PERIODIC (statement of account plus due re-probes). Probe UNDERSTANDING of the load-bearing step, never RECALL of incidentals. Do NOT use to generate practice (concept-exercise-generator) or teach a concept fresh (professor-mentor-technical-teaching).

0
installs
▌
Research Proposal Writer · jurgendn bundle
Write or refine a forward-looking research proposal — a plan for work not yet done, judged on plausibility and feasibility rather than results. Two modes: a PhD research proposal for a prospective supervisor, admissions committee, or fellowship (what you will do over the next several years), or a next-paper proposal pitched to your current advisor (what the next paper is and how you will get there). Trigger on "refine my research proposal", "research plan for my PhD application", "proposal for my professor about the next paper", "what I'll work on for the next N years", "pitch my next project to my advisor", "is my research plan convincing". NOT the SOP essay (apply-sop-writer), not grading a finished draft against a named reader (professor-critic), not stress-testing the idea itself (research-idea-stress-test), and not building defense slides (deck-beamer-proposal-report).

0
installs
▌
Theory Counterexample Hunter · jurgendn
Search for edge cases, pathological constructions, and violated assumptions that break a conjecture or weaken a theorem. Use when a claim feels too broad, an assumption seems suspiciously weak, or a proof step needs stress-testing. Trigger on phrases like "is this always true", "does this hold in general", "check this claim", "try to break this", "stress-test this theorem", "find a counterexample", "is this assumption necessary", or whenever the user states a result and asks whether it can be trusted. Also trigger when reviewing a proof and a step feels hand-wavy or a quantifier feels wrong.

0
installs
▌
Abstract And Intro Writer · jurgendn
Draft or revise the abstract and introduction of a research paper. Use whenever the user needs to write or improve these sections — including phrasings like "write my abstract", "improve my intro", "draft the introduction", "my abstract is weak", "how should I open the paper", "the intro is too long", "I need a hook", "write a one-paragraph summary", or "my introduction doesn't sell the contribution". Also use when the user has a paper plan or results but no prose yet for the opening. The abstract and introduction have distinct structures and reviewer conventions that differ from the rest of the paper; this skill handles both as a pair since the intro must match and extend the abstract's promises. If the contribution or paper spine is not yet stable, use paper-argument-planner first.

0
installs
▌
Reviewer Response Strategist · jurgendn bundle
Plan rebuttals, reviewer responses, and revision strategies for research papers. Use this skill whenever the user shares peer reviews, reviewer concerns, meta-review text, acceptance/rejection feedback, or asks how to respond to Reviewer 2, what experiments to run for rebuttal, what to concede, how to phrase a response, or how to prioritize limited rebuttal space. This skill focuses on strategy and response structure; use results-writeup for writing new result prose after the strategy is decided.

0
installs
▌
Model Eval Error Analysis · jurgendn
Evaluate a model rigorously and inspect where it fails. Use after training, during comparison to baselines, for LLM/system evaluations, noisy or suspiciously clean results, or whenever benchmark numbers alone are insufficient and the user needs slices, failure categories, and explanations.

0
installs
▌
Statistical Testing Guide · jurgendn bundle
Choose the right statistical test, compute effect sizes, design power analyses, and report statistics correctly in ML and AI research papers. Use whenever the user asks how to compare two models statistically, whether results are significant, how many seeds to run, how to report variance, which test to use for non-normal data, how to handle multiple comparisons, or how to answer a reviewer challenging statistical validity. Also use when planning an evaluation and needing to decide the number of seeds/trials before running, or when results are borderline and the user needs to know if they can claim significance. Prevents the most common statistical mistakes in ML papers: reporting only means without variance, choosing seeds post-hoc, and conflating statistical significance with practical significance.

0
installs
▌
Publication Grounded Solution Design · jurgendn bundle
Design industry R&D solution directions grounded in prior successful cases, academic papers, technical reports, and credible industry publications. Use whenever the user has a vague business, banking, fintech, enterprise, or operational problem and wants to brainstorm methods, draft a theoretically motivated solution, compare solution families, or justify a proposal with publications instead of unsupported intuition. Also use when the user asks for "prior successful cases", "paper-backed solutions", "method draft", "research-backed proposal", or "what has worked before" for an applied industry problem.

0
installs
▌
Discussion Report Synthesizer · jurgendn
Synthesize a fragmented academic conversation into a cohesive, structured markdown report. Use this skill whenever the user has asked many disconnected questions about a topic in one session and now wants a unified write-up that captures what was discussed. Trigger phrases: write a report on our conversation, summarize this discussion into a document, turn this into a readable report, capture everything we talked about, make a note of this whole conversation, write up what we covered. The skill scans the full conversation, groups ideas into themes, identifies gaps, asks 2–4 targeted clarifying questions to fill them, then drafts the report. It does NOT write experimental results sections (use results-writeup), Related Work prose (use related-work-writer), or paper section drafts (use method-section-writer or abstract-and-intro-writer).

0
installs
▌
Paper Idea And Scope Brainstormer · jurgendn bundle
Brainstorm a research idea and decide what a new paper should cover — generate candidate framings, pick the central contribution, and draw the in-scope vs. deferred boundary. Use whenever the user has a rough interest, preliminary results, or a problem and asks "what should this paper be about", "help me brainstorm an idea", "what should I include / cover", "is this one paper or two", "help me scope this", "what's the contribution here", or "I have an idea but don't know how to frame it". Default to running this upstream of research-idea-stress-test (which vets one chosen idea) and paper-argument-planner (which structures a paper once the contribution is fixed). Always check first whether the input is sufficient (a seed, the field, the motivation); if not, request the missing context before brainstorming rather than inventing a direction. Hands off per-claim evidence to benchmark-and-baseline-selector, novelty grounding to literature-triangulation, and section logic to paper-argument-planner.

0
installs
▌
Theory Paper To Theorem Distiller · jurgendn
Distill a dense theory paper into definitions, main claims, proof strategy, and reusable techniques. Use for theoretical ML or math paper reading, seminar prep, related-work synthesis, extracting reusable lemmas, or deciding what part of a paper is actually load-bearing.

0
installs
▌
Concept Exercise Generator · jurgendn bundle
Generate a graded exercise set — easy to advanced — that lets a learner verify whether they actually understand a topic, concept, method, derivation, or paper, with solutions kept in a separate file for honest self-testing. Use whenever the user says "give me exercises", "quiz me on X", "test whether I actually get this", "problems from easy to hard", "practice questions with solutions", "how do I check if I understood this", or wants a self-check after learning something. Produces two artifacts: an `exercises.md` (problems only, tagged by difficulty tier and concept) and a `solutions.md` (worked answers plus a self-check rubric and a pointer to what to revisit on a miss). This skill builds the verification material; it does not teach the concept (use professor-mentor-technical-teaching for that) and it does not sequence a learning path (use flow-learn-new-topic for that). It is one of the assessment steps those flows hand off to at an exit gate (the other being a discussion round).

0
installs
▌
Apply Recommendation Letter Strategist · jurgendn bundle
Plan PhD and research-application recommendation letters: choose recommenders, assess letter strength and risk, draft request or reminder emails, and prepare recommender packets. Use this skill whenever the user asks who should write their letters, whether a recommender is strong enough, how to ask for a recommendation, what materials to send, or how to coordinate letters for PhD programs, fellowships, research internships, RA roles, or lab applications. Establishes the target's declared theme or track from its live official page first, since it decides which recommender's observed evidence matters most and what the packet must supply.

0
installs
▌
Benchmark And Baseline Selector · jurgendn bundle
Given a research proposal, claim, method, or result, recommend the baselines and benchmarks needed to support it — split into a MINIMAL set (necessary for credibility, skipping any risks rejection) and a SUGGESTED set (complementary, strengthens external validity). Use whenever the user proposes something and asks "what should I compare against", "what baselines do I need", "which benchmarks support this claim", "is this enough to be convincing", "what's the minimal experiment set", "what would a reviewer demand", or shares a method/idea and wants the comparison plan. Default to running this on any proposal so claims are backed by the right evidence. Always check first whether the input is sufficient (precise claim, field, the novel component, metric); if not, request the missing context before recommending rather than guessing. Use experiment-design to design the run itself (or to design a measurement when no benchmark exists), and hypothesis-and-ablation-planner for internal ablations.

0
installs
▌
Hypothesis And Ablation Planner · jurgendn
Design ablations that isolate causal mechanisms instead of producing decorative tables. Use during paper preparation, benchmark analysis, reviewer response, or whenever multiple mechanisms could explain an observed gain and the user needs to determine which component matters, why it matters, and whether alternative explanations remain plausible.

0
installs
▌
Professor Mentor Technical Teaching · jurgendn bundle
Explain advanced technical topics the way a senior professor mentors a graduate student — intuition and motivation first, mathematical rigor second, critical judgment throughout. Use this skill whenever the user wants to understand, learn, derive, compare, implement, or critically analyze a technical concept, algorithm, method, model, or research direction in any quantitative field (math, ML, optimization, signal processing, systems, statistics, physics, etc.). Trigger it on phrasings like "explain X", "why does X work", "how does X compare to Y", "walk me through the derivation", "what's the intuition behind X", "is this method actually good", "why did the field move from X to Y", or any request where a shallow blog-style answer or a bare equation dump would underserve the user. Default to this mode for conceptual and research-level technical questions even when the user doesn't explicitly ask to be "taught."

0
installs

88 published skills