Claude

Fetches preprints posted to bioRxiv or medRxiv within a given date window, then keyword-filters the results client-side. Wraps the public `api.biorxiv.org/details/{server}/{from}/{to}/{cursor}` endpoint, handles cursor pagination, normalizes records to a stable shape (doi, title, authors, abstract, date, server, version, url), and applies a keyword-OR match against title + abstract. Domain-neutral — usable for any biology / clinical preprint scan, not just one project. Use when user mentions bioRxiv, medRxiv, weekly preprint scan, fetch preprints, last-N-days preprints, or when a literature-scan agent needs structured preprint records.

Combines Pareto prioritization (80/20), timeboxing, and deep work techniques to manage attention, eliminate context-switching, and maximize high-impact output. Use when managing time and attention, combating procrastination, prioritizing high-impact work, planning daily/weekly schedules, or when user mentions timeboxing, Pomodoro, deep work, 80/20 rule, Pareto principle, focus blocks, task batching, or energy management.

Moves prose deliberately between the concrete and the abstract and stops it stranding in the middle rung where bureaucratic prose lives. Tags every paragraph R1 (named particulars) to R4 (meaning), enforces run rules, detects the middle-rung trap, repairs it by climbing down to a real particular rather than simplifying vocabulary, gates the top rung with a counter-sentence test, and closes with an A-frame descent. Use when a draft reads as dense, vague, or jargon-heavy, when a readability score fails, when a conclusion feels grandiose or unearned, when technical material will not land for a general reader, or when user mentions ladder of abstraction, too abstract, too vague, concrete example, middle rung, dead zone, jargon, altitude, zoom in, zoom out, and thus capitalism.

Rewrites a published substacker essay as a LinkedIn post with a hook fitting the 210-char fold, practitioner framing (less confessional than Substack), short 2-3 line paragraphs, and 0-2 niche hashtags. 900-2500 characters. Emits linkedin-post.md. Use as the LinkedIn-native arm of the Distribution Translator. Trigger keywords — LinkedIn post, LinkedIn rewrite, practitioner, professional network, niche hashtags.

Weekly refresh of per-opponent archetype + behavioral profiles for the 11 opposing teams in the user's Yahoo Fantasy Baseball league (ID 23756). Thin baseball-specific wrapper around the domain-neutral `opponent-archetype-classifier` -- provides the 10-archetype MLB taxonomy (balanced, stars_and_scrubs, punt_sv, punt_sb, punt_wins_qs, hitter_heavy, pitcher_heavy, inactive, frustrated_active, unknown), extracts MLB features from Yahoo pages (draft distribution, FAAB spend, waiver pattern, roster composition, lineup consistency, trade activity, recent record, activity recency), invokes the classifier, and writes/updates `context/opponents/<team-slug>.md` files per `opponent-profile-schema.md`. Read-modify-write preserves manual notes. Emits a weekly summary signal at `signals/wkNN-opponent-profiles.md`. Use when user says "opponent profiling", "classify opposing manager", "update opponent profiles", "refresh opponents", "weekly opponent scout", or "MLB fantasy opponent archetype".

Counts MLB games per team during the Yahoo fantasy playoff window (weeks 21, 22, 23 -- Aug 17 through Sep 6, 2026) and grades the quality of each team's opponents. Emits three signals per rostered player -- playoff_games (int, max ~21), playoff_matchup_quality (0-100), holding_value (0-100) -- that drive trade-deadline and playoff-lineup decisions. Use when the user mentions playoff weeks, weeks 21-23, playoff schedule, game count, holding value, or asks whether to keep/trade a player for the playoff run. Pre-July 1 this skill returns "insufficient signal -- too early"; from July 1 onward it fires weekly.

Turns a point-estimate valuation into a distribution — samples the drivers you are least sure about from normal, lognormal, triangular, uniform or discrete-scenario distributions, re-runs the DCF engine on every draw, and reports percentiles, mean, standard deviation, the probability the value exceeds the market price, and the share of draws the engine refused as infeasible. Correlates drivers through a shared common factor. Use when running a Monte Carlo simulation or a probabilistic valuation, putting a range or confidence band around a value per share, asking how likely it is that a stock is under- or overvalued, valuing a commodity or cyclical company where one macro variable such as the oil price dominates, or turning a scenario grid into a probability-weighted expected value.

Maps researched evidence onto Truby's twenty-two structural steps without fabricating the steps the evidence cannot fill. Runs the designing-principle discrimination test, step-fit triage (PRESENT / ABSENT / NOT APPLICABLE), per-slot warrant cards, the empty-slot decision table, the moral-argument spine, and the revelation-intensity audit, then stops at a tagged scene weave. Use when architecting a long-form nonfiction piece from a research corpus, structuring a post-mortem, market history, biography, or science writeup, or when user mentions Truby, 22 steps, designing principle, moral argument, revelation sequence, scene weave, or step-fit triage.

Decides which narrative form a body of researched material can actually support, before any structure is imposed. Runs a six-question form-triage gate producing STORY NARRATIVE, EXPLANATORY NARRATIVE, MULTI-NODE GATHERING, or DO NOT NARRATE; runs the ABT (And/But/Therefore) spine diagnostic; selects a spine from a thirteen-structure table covering Truby arc, layer cake, pyramid/SCQA, OCAR, PR/FAQ, sparkline, Raskin five-slot, martini glass, drill-down, inverted pyramid and braided/mosaic; and writes an immutable spine contract with a boredom budget. Use before outlining or drafting any long-form nonfiction, report, post-mortem, market history, research writeup or case study, or when user mentions find the narrative, what's the story here, narrative arc, story structure, how should I structure this, does this have a story, or storytelling with data.

Identifies, assesses, and prioritizes project risks using probability-times-impact scoring (risk matrix), then assigns owners, mitigation plans, contingencies, and triggers to track risk evolution over the project lifecycle. Use when managing project uncertainty, building a risk register, conducting risk assessments, defining risk mitigation plans, or when user mentions risk register, risk management, probability-impact matrix, or asks "what could go wrong with this project?".

Systematically identifies vulnerabilities, threats, and mitigations for systems handling sensitive data using STRIDE methodology, trust boundary mapping, and defense-in-depth principles. Use when designing or reviewing systems with PII/PHI/financial/auth data, building security-sensitive features (auth, payments, file uploads, APIs), preparing for audits or compliance (PCI, HIPAA, SOC 2), investigating incidents, or integrating third-party services. Use when user mentions threat model, STRIDE, trust boundaries, attack surface, or security review.

Rewrites a published substacker essay as a Substack Note using the extracted spine and chosen hook. Closest voice to the essay. Bolded maxim closer. Single link line. 60-180 words. Emits substack-note.md in the post's distribution folder. Use as the Substack-native arm of the Distribution Translator. Trigger keywords — Substack Note, note rewrite, note post, tease, Notes feed.

Synthesizes information from multiple sources into coherent insights and applies analogical reasoning to transfer knowledge across domains. Use when conducting literature reviews, integrating stakeholder feedback, reconciling conflicting viewpoints, identifying cross-source patterns, creating explanatory analogies ("X is like Y"), finding creative solutions through cross-domain transfer, or testing whether analogies hold (surface vs deep). Use when user mentions "synthesize", "combine sources", "analogy", "similar to", "transfer from", "integrate findings".

Domain-neutral methodology for evaluating completeness and logical soundness of an extracted set of components, then transforming them into actionable guidance. Runs the "is it true / is it complete / what of it" critical evaluation pass before any final artifact is built. Checks for completeness gaps, logical consistency, contradictions, and practical applicability. Reusable across any extraction workflow - skill creation (evaluating extracted components before building SKILL.md), paper extraction (evaluating Pass 2 extraction notes before deep reading), report writing (evaluating gathered evidence before synthesis). Use when an agent has extracted structured components from a source and needs to gate-check before downstream commitment. Trigger keywords - synthesis evaluation, completeness check, logic check, critical evaluation, fact-check before synthesis, gap analysis, what is not said.

Appends one structured YAML observation block to substacker shared-context/audience-notes.md iff the week produced at least one observation with confidence ≥ medium. Includes supporting evidence (post slugs + numbers) and reviewed_by_curator flag. Never rewrites or deletes prior entries. Append-only discipline protects downstream agents' shared context. Use at the end of each weekly pipeline after write-weekly-report. Trigger keywords — audience notes, append observation, audience insight, confidence-rated.

The mutation operator for the FIFA World Cup Fantasy evolution engine. Applies small, bounded, feasibility-preserving perturbations to a recombined offspring squad/plan — swap one differential for an adjacent-price alternative, reorder the bench by kickoff, shift one captain-ladder step, try one fixture-swing transfer, nudge one enabler. Mutation rate is low by default and raised on demand by the diversity governor when the population has collapsed. Each mutation stays inside budget/nation/formation feasibility. Use in the mutate stage after recombination, or when wc-population-diversity calls for more exploration.

Composes the final Technical Reviewer artifact at ops/technical-reviewer/YYYY-MM-DD-{slug}-review.md. Enforces frontmatter schema, section order (Summary → Blockers → Claims → Boundary-Break Suggestions → Glossary Alignment → Could-Not-Verify → Research Log), go/no-go decision rule, and never-modify-draft principle. Use exactly once per Technical Reviewer run as the last step. Trigger keywords — write review, technical review artifact, compose review, claim review output.

Crafts the one-sentence promise a substacker section makes to its reader — specific, testable, non-overlapping with other sections, written in the writer's voice (not marketing). Use when propose-section stages a new section or when an existing promise is being revised. Trigger keywords — section promise, one-sentence promise, section statement, reader promise.

Checks whether the substacker corpus has enough material to justify a Curator run. Counts published posts, time-gates against the last review, and reports go/no-go with the specific gate that failed. Use before any Curator run, and on cold start to decide whether to propose sections yet. Trigger keywords — readiness, corpus ready, gate check, cadence gate, pre-flight.

For each Trend Scout candidate item, checks substacker shared-context/topic-ledger.md and tags with NEW | OVERLAPS seed:{slug} | OVERLAPS draft:{slug} | OVERLAPS published:{slug}. Adds a reinforcement_angle note for items overlapping with published posts ("external confirmation of X"). Read-only against the ledger. Use after summarize-signal, before rank-by-user-fit. Trigger keywords — cross-ref, ledger check, overlap, reinforcement, dedup external.

Designs complete GraphRAG systems integrating graph databases, vector stores, orchestration frameworks, and LLM reasoning. Guides through pattern selection, technology stack decisions, integration pipeline design, and domain-specific customizations. Use when designing GraphRAG systems, choosing technology stacks for graph-augmented retrieval, combining Neo4j with LLM, using LangChain/LlamaIndex knowledge graphs, applying community detection for RAG, building hybrid symbol-vector pipelines, or deploying production or domain-specific GraphRAG.

Produces an explicit component-by-component mapping from the analogy's source domain to the target technical concept. Rejects vague analogies by forcing each source element to map to a specific target element, and flags unmapped elements as voice-breaking ("it's like a brain" is rejected because "brain" is unmapped). Use after generate-analogy-set, for each of the 5 framings. Trigger keywords — map, component mapping, source target, explicit mapping, what does the X correspond to.

Identifies fantasy baseball players whose surface stats (wOBA, ERA, batting average) are diverging from their underlying Statcast quality (xwOBA, FIP, xBA) — emits a `regression_index` from -100 (very lucky, sell high) to +100 (very unlucky, buy low). Primary signal for buy-low/sell-high decisions on trades and waivers. Use when user mentions "buy low", "sell high", "regression candidate", "lucky", "unlucky", "xwOBA gap", "ERA-FIP gap", "BABIP", "due for regression", or is deciding whether to trade for / trade away a player based on over- or under-performance.

Groups a set of kept papers into 2-5 thematic clusters before synthesis, using abstract semantics + matched-keyword overlap. Names each cluster with a short noun phrase that describes what the cluster argues, not just the topic. Surfaces an "outliers" bucket for single-paper themes that don't fit. Domain-neutral - usable for any literature-scan workflow. Use after relevance filtering and before writing the synthesis report. Trigger keywords - cluster papers, group by theme, thematic clusters, paper themes, organize papers.

Scores a candidate paper against a keyword watchlist and a relevance-criteria document, returning KEEP/DROP/REVIEW with a one-line rationale and a 0-100 relevance score. Combines keyword-match strength, criteria-fit, and a historical-context check (was this paper or its preprint already covered in a recent digest). Domain-neutral - usable for any literature-scan workflow. Use after fetching candidate papers from bioRxiv, medRxiv, or PubMed and before clustering or synthesis. Trigger keywords - paper relevance, filter papers, keep or drop, score papers, relevance rationale.

Checks paragraph rhythm in a substacker draft — long/short mix, one-sentence paragraph at pivots, no walls, avoid monotony. Flags drafts where >3 consecutive paragraphs share the same length bucket, where the pivot lacks a one-sentence paragraph, or where any paragraph exceeds 120 words. Use in the Editor's structural pass. Trigger keywords — rhythm, paragraph length, wall of text, one-sentence paragraph, pivot, monotone.

Assesses whether a company returns the right amount of cash to shareholders. Computes FCFE over a history under three debt assumptions, compares it against dividends plus buybacks to give the cash surplus or deficit, places the firm in the dividend matrix of affordability against project quality, projects forward FCFE and buyback capacity, and benchmarks payout ratio and dividend yield against peers and the market. Handles bank FCFE from regulatory capital. Use for dividend policy, payout ratio, dividend yield, potential dividends, FCFE, cash returned, buybacks, share repurchase capacity, excess cash, dividend cut or increase, cash accumulator or overpayer, or the dividend matrix.

Produces the counterfactual framing in an Intuition Builder 5-set — "what if this component were not here?" Reveals the function of a technical element by subtracting it and observing what breaks. Uses Pearl's causal ladder (counterfactual = level 3) as the theoretical spine. Use as the 5th archetype slot of generate-analogy-set, or invoked standalone when the writer wants to build intuition for why a specific element exists. Trigger keywords — counterfactual, what if not, remove, subtract, reveal function, why does this exist.

Runs Jack Hart's Wordcraft line passes in strict order (Structure, Force, Brevity, Clarity, Rhythm, Humanity, Color, Voice, Mechanics) followed by a slop audit, one pass per edit class, with a claim-strength invariant that stops verb strengthening and hedge cutting from silently overclaiming. Enforces upward escalation instead of downward compensation, a protected-hedge list, a clarity split log the rhythm pass may not undo, and rhythm metrics reported as diagnostics rather than optimized. Use when line-editing a draft, tightening flabby prose, fixing weak verbs or monotonous sentence rhythm, de-slopping AI-sounding copy, or when user mentions line edit, force pass, punch it up, tighten this, passive voice, wordy, sentence variety, sounds like AI, burstiness, or Wordcraft.

Appends an entry to substacker shared-context/analogy-catalog.md when the writer PUBLISHES a post that uses a new analogy. Not invoked on seed or draft — only on publish. Records source, target, post, freshness, mapping, where-it-breaks, and why-it-worked. Prevents silent recycling in future Intuition Builder runs. Use at publish time for any post that contains a non-trivial analogy. Trigger keywords — catalog, analogy catalog, update catalog, publish, analogy archive.

Rates fixture difficulty (separately for an attacking and a defending fantasy asset, 1-5 each way, from opponent xGA/xG), computes team progression probabilities (group qualification odds from the mini-table; round-by-round knockout survival down the bracket) by reference-class forecasting updated Bayesianly on observed form and group state, detects weak-group/mismatch fixtures (powerhouse-vs-minnow -> captain candidates and clean-sheet locks), and builds a multi-round swing calendar of where each team's difficulty flips so transfers can be planned ahead. Emits the `fixture` signal. Use when called by wc-fixture-analyst, by the strategists in the populate stage, and by wc-fitness-eval for its progression_carry term — and never re-derive these once the signal exists.

Produces a Bayesian prior probability that an offered transaction is +EV for the recipient, given that the counterparty chose to propose it. Applies Akerlof market-for-lemons logic -- if they offered it, they believe it is +EV for them, so the prior that it is +EV for us is materially below 50%. Reusable across trade evaluation, waiver drops (another team dropping a player is also adverse selection), job-offer analysis, M&A, and any "someone offered me this" situation. Use when you receive an unsolicited trade/offer/proposal, analyzing incoming trade prior, evaluating why a counterparty proposed a deal, or when user mentions adverse selection, market for lemons, why did they offer this, incoming trade prior, they proposed it, Bayesian adjustment on received offer.

For each spending category, computes current-period spend, 6-month rolling average, year-over-year delta, and budget variance, then flags categories that are outliers (>1.5x rolling average or >130% of budget). Produces a ranked list of categories that grew, shrank, or stayed flat, plus the top transactions driving each outlier. Use for monthly spending reviews, identifying lifestyle creep, evaluating budget adherence, or when user mentions category trends, spending changes, budget variance, or outlier spend.

Chains together clear specifications, proactive risk analysis (premortem/register), and measurable success metrics into a comprehensive planning artifact for high-stakes initiatives. Use when planning migrations, launches, or strategic changes that need implementation roadmaps, risk mitigation, and instrumentation. Invoke when user mentions "plan this migration", "launch strategy", "implementation roadmap", "what could go wrong", "how do we measure success", or when high-impact decisions need comprehensive planning.

Performs axial-coding-style thematic clustering over the substacker corpus of published posts to surface candidate sections. Uses Braun & Clarke's six-phase thematic analysis — familiarization, initial coding, searching for themes, reviewing themes, defining themes, naming. Reads full bodies, not titles. Use when re-opening the section question. Trigger keywords — cluster, theme, axial coding, thematic analysis, candidate sections.

Parse a heterogeneous conference program (markdown, HTML, PDF-derived text, or JSON) into normalized event records with per-field confidence scores and independent classification axes (topic, depth, format, prerequisites, recorded, capacity). Detects the program's format before extracting, treats every inferred field as uncertain (present vs inferred vs missing), and flags thin or missing abstracts so downstream enrichment can target them. Conference-agnostic. Use when ingesting a conference or event schedule into a structured store, normalizing a talk/session list, or extracting per-session metadata with calibrated confidence. Trigger keywords - program ingestion, parse schedule, session extraction, event records, conference program, talk metadata, per-field confidence.

Estimates discount rates and the optimal financing mix — riskfree rate in any currency, equity risk premium including country risk, bottom-up and total betas, synthetic credit ratings from interest coverage, cost of debt, market value of debt, WACC, the implied equity risk premium solved from current index levels, and the cost-of-capital schedule across debt ratios that locates the optimal debt ratio, with an APV cross-check and a stress test under rating constraints. Bundles Damodaran's rating and spread tables, Altman default probabilities, industry-average betas and multiples, and country risk premiums, each tagged with its vintage. Use when estimating a discount rate, WACC, hurdle rate, cost of equity or cost of debt, unlevering or relevering a beta, when a company has no credit rating, when estimating the equity risk premium priced into the market, when looking up a sector beta or country risk premium, when converting a rate between currencies, or when finding or stressing an optimal capital structure.

Systematically evaluates existing designs against cognitive science principles using repeatable checklists, scoring rubrics, and severity-classified fix recommendations. Use when conducting design reviews or critiques, evaluating designs for cognitive alignment, performing quality assurance before launch, diagnosing usability issues, or choosing between design alternatives with objective criteria.

Uses WebFetch to pull publicly visible subscriber count and per-post public view count from substacker's Substack archive page and individual post URLs. Supplements the CSV when subscriber-count field is stale (>24h old) or when a post has public shares not yet reflected. Rate-limited to ≤10 fetches per invocation. Use when CSV subscribers-end field may have drifted or when external-share attribution needs a public signal. Trigger keywords — public stats, Substack public page, subscriber count check, post views supplement, WebFetch.

Fetches the last 7 days of updates from every entry in the substacker Trend Scout watchlist — blogs, paper aggregators (arXiv, Hugging Face papers), social feeds. Returns normalized {title, url, author, published, excerpt, source_type} tuples. Use at the start of a weekly Trend Scout run. Trigger keywords — watchlist, fetch sources, weekly fetch, last 7 days, source normalization.

Maps every input a company analysis needs to where it comes from — filing line items, market data, Damodaran reference datasets, macro series — with units, update frequency, acceptable fallbacks and the consistency rules that bind them. Use when gathering data for a valuation, when an input is missing and a substitute is needed, when deciding whether a user-supplied figure can be accepted at face value, or when checking that figures are internally consistent.

Transforms vague or unreliable prompts into structured, constraint-aware prompts with explicit roles, task decomposition, output formats, and quality checks. Use when prompts produce inconsistent outputs, need explicit structure and constraints, require safety guardrails, involve multi-step reasoning that needs decomposition, need domain expertise encoding, or when user mentions improving prompts, prompt templates, structured prompts, prompt optimization, reliable AI outputs, or prompt patterns.

Converts baseball and fantasy-baseball jargon into plain English for a user with zero baseball knowledge. Wraps every user-facing sentence produced by the MLB agent team (morning briefs, trade recommendations, waiver calls, chat summaries). Detects jargon terms, attaches an inline parenthetical plain-English gloss on first mention in a document, enforces the action-verb ladder (START / SIT / ADD / DROP / BID $X / ACCEPT / COUNTER / REJECT), and rejects assumed-knowledge phrases like "hot streak" or "positive matchup." Use when asked to translate for beginner, explain in plain English, translate this, write without jargon, make it beginner-friendly, or produce any user-facing MLB output for K L D'Souza's Fantasy Baseball 2K25 team.

Parses Yahoo Fantasy Baseball league state (roster, standings, current matchup, FAAB remaining, free agents) from authenticated Yahoo team pages via Claude-in-Chrome browser automation, then grounds it against league-config.md and team-profile.md to emit a normalized league-state bundle every other agent can consume without re-scraping. Use when the coach or any downstream agent needs to read Yahoo roster, refresh team profile, pull league state, get current matchup, check FAAB remaining, list free agents, or when user mentions "what's on my roster", "who am I playing this week", "how much FAAB do I have left", or "refresh my team".

Stops a true set of facts from being assembled into a false story. Runs the retrospective slot audit, the outcome-blind rewrite (flip test, separation test), the inevitability audit with its overshoot check, the survivorship graveyard pass, the counterfactual admissibility gate, and the expensive proportion and omission audits. Every check labels and discloses; none deletes. Use when drafting or reviewing history, post-mortems, biography, market or protocol narratives, research writeups, or any account written after the outcome was known, or when user mentions hindsight bias, narrative fallacy, survivorship bias, inevitability, halo effect, outcome bias, just-so story, teleology, or Whig history.

Guides validation of ideas before full development using pretotyping (fake doors, concierge MVPs, Wizard of Oz) and prototyping at appropriate fidelity (paper, clickable, coded) to test assumptions about demand, pricing, and feasibility. Use when testing ideas cheaply before building, choosing prototype fidelity, running experiments to validate assumptions, or when user mentions prototype, MVP, fake door test, concierge, Wizard of Oz, landing page test, smoke test, or asks "how can we validate this idea before building?".

Builds scenes that are reported rather than generated, by running a four-slot SCAM qualification gate (Setting, Character, Action, Meaning) with a source required per slot, a scene floor and ceiling, the 1-of-20 telling-detail cut with a provenance veto, the RUE pass, and a scene/summary narrative-distance audit against per-form ratio targets. Every technique has a person variant and a system variant (market, protocol, institution, codebase, supply chain). Use when turning research notes into a scene, opening a section with a moment, deciding whether a passage is a scene or summary, or auditing whether a vivid paragraph is actually sourced, or when user mentions scene, SCAM, show don't tell, telling detail, scene vs summary, summary in costume, establishing shot, make this vivid, or in medias res.

Computes a 0-10 intuition-density score for a seed body using 8 concrete measurable signals — analogy presence, concrete worked example, counterfactual offered, reframe against default, biology-to-AI transfer, question posed, calibrated hedge, math-to-metaphor handoff. Emits both the numeric score and the list of triggered signals for auditability. Use after topic tagging to enrich seed frontmatter in the substacker Librarian pipeline. Trigger keywords — intuition density, score, signals, analogy count, worked example, density proxy.

Provides frameworks for mapping stakeholder influence networks, designing team structures aligned with system architecture (Conway's Law), defining team interface contracts (APIs, SLAs, decision rights), and assessing capability maturity (DORA, CMMC, agile models). Use when designing org structure or team topologies, mapping stakeholders for change initiatives, defining team interfaces, assessing capability maturity, planning restructures, or when user mentions org design, team structure, stakeholder map, Conway's Law, or RACI.

Assigns a category and subcategory to a financial transaction by matching its raw description against a configurable taxonomy and rules table, falling back to LLM inference when no rule matches. Emits a normalized merchant name, category path, recurring flag, and confidence score, and proposes new rules from confirmed classifications. Use when categorizing bank, credit-card, or brokerage transactions, building or refining a category taxonomy, or when user mentions transaction categorization, merchant normalization, expense classification, or category rules.

Measures and protects diversity in the FIFA World Cup Fantasy evolution engine — the anti-inbreeding / selection-pressure governor from the Evolution document. Computes how collapsed an offspring set is (pairwise squad overlap %, captain overlap, ownership-profile spread, variance-band coverage); if the population has converged toward one template (premature convergence / local optimum), it injects an under-represented genotype's blocks and/or raises the mutation rate and signals a re-run; and it tunes selection pressure (too high collapses the gene pool, too low makes the board noise). Ensures the manager's board always offers a real choice across the variance spectrum. Use after recombination + mutation, before emitting offspring.

Assigns a substacker draft or published post to the best-fitting section (or to unassigned) based on content + section promises in section-map.md. Used by the Editor on every draft review (to load the right voice overlay) and by the Curator in batch mode. Trigger keywords — classify post, section assignment, which section, route post, per-draft section.

Guides a learner to invent a math or ML concept themselves through a Socratic walk — a sequence of small guessable questions that ends with the learner stating the formal definition unprompted. The 3Blue1Brown signature move. Use when the learner is meeting a foundational concept (eigenvectors, gradient, attention, softmax, KL divergence) for the first time, when prior exposure produced memorization without understanding, or when the user says "explain it from scratch", "I want to really get it", "build it up for me", or "where does this come from".

Thicken a thin or missing conference-session abstract through targeted web research so short, noisy text becomes classifiable — pulling the speaker's recent work, the linked repo or paper, and the company/product to add signal. Every enriched claim carries provenance (source URL + retrieval date) and a confidence, the enriched text is stored separately from the original (never overwriting it), and enrichment is gated so already-rich abstracts are left alone. Conference-agnostic. Use when an event record has been flagged thin by program extraction, when a session lacks enough text to classify, or when a key axis (depth, topic) is low-confidence. Trigger keywords - abstract enrichment, thicken abstract, enrich session, speaker research, thin abstract, runtime web search for talks.

Analyzes profitability per customer, product, or transaction to determine business model viability and scalability. Covers CAC, LTV, contribution margin, cohort analysis, and growth-readiness assessment. Use when evaluating business model viability, validating startup metrics (CAC, LTV, payback period), making pricing decisions, comparing business models, or when user mentions unit economics, CAC/LTV ratio, contribution margin, customer profitability, or break-even analysis.

For each technical term in a substacker draft's claims, checks shared-context/glossary.md for a writer-specific definition and compares to field-standard. Emits a note for terms where the two diverge (aligned / diverged-safe / diverged-risky). Feeds into classify-claim (which definition governs) and write-review-artifact's Glossary Alignment section. Use on every claim. Trigger keywords — glossary alignment, term definition, writer vs field, load-bearing term.

Guides creation and review of competitive grant proposals (NIH R01/R21/K, NSF, foundations) by applying reviewer-perspective thinking to ensure clear hypotheses, compelling significance, genuine innovation, and feasible approaches. Use when writing or reviewing grant proposals, crafting specific aims, drafting significance/innovation/approach sections, or when user mentions R01, R21, K-series, grant writing, proposal review, study section, or fundable hypothesis.

Organizes, structures, and labels content so users can find and manage information effectively. Guides through content audits, card sorting, taxonomy design, navigation structure, and tree testing validation. Use when organizing content for digital products, designing navigation systems, restructuring information hierarchies, improving findability, creating taxonomies or metadata schemas, or when users mention information architecture, IA, sitemap, navigation design, content structure, card sorting, tree testing, taxonomy, or findability.

Establishes pre-defined, objective conditions for stopping projects and specific decision points for continue/pivot/kill evaluations. Guides through defining kill criteria, setting go/no-go gates, avoiding sunk cost fallacy, and executing disciplined stopping decisions. Use when defining stopping rules for projects, avoiding sunk cost fallacy, setting objective exit criteria, deciding whether to continue/pivot/kill initiatives, or when users mention kill criteria, exit ramps, stopping rules, go/no-go decisions, project termination, or sunk costs.

House style for learning-in-public essays — a curious practitioner thinking out loud while learning a hard technical subject (here, ML-driven crop genetics / genomic selection). First-person, concrete-first, honest about the edge of understanding, mechanism over vocabulary. Provides the register, the hook patterns, the anti-slop hard rules, and the rule for reading a per-writer voice-profile.md so the voice stays the writer's own. Use when drafting a vault-style post from evergreen notes, or as the lens an advisory editor critiques against. This skill never imposes voice; it describes defaults and defers to the writer's profile.