Data & Analytics for Claude Code

Weekly Substack performance analyzer for substacker. Pulls stats directly from the Substack dashboard via Claude-in-Chrome browser automation (primary) or ingests a manual CSV export (fallback). Computes rolling 4-week baseline, attributes over/under-performance in plain English with calibrated confidence, tracks per-section metrics once sections exist, produces a 400-800 word weekly report. Feeds Curator (section pruning) and Growth Strategist (quarterly rollups). Use Monday mornings. Trigger k

Generates the weekly static HTML dashboard for the household finance JSON store. Reads the canonical data files, applies cognitive-design principles for visual hierarchy and visual-storytelling-design for narrative annotations, runs cognitive-fallacies-guard to prevent visual misleads, and produces a single self-contained HTML file with six panels (net worth over time, cash flow, recurring & subscriptions, goals, asset allocation, vigilance feed). Output is a static file — no server, no live dat

Household spending analyst that answers, every month, where the money went, what changed versus prior months and same-month-last-year, what is recurring, what is an outlier, and what is coming up. Produces the Spending section of the monthly briefing, runs a quarterly subscription audit, projects 60-day daily cash flow per account, and maintains a 12-month forward seasonal calendar. Use as the third phase of a per-drop or monthly pipeline, when the user asks for a spending review, or when sizing

Search worker for one keyword query across bioRxiv, medRxiv, PubMed, and arXiv. Spawned by an orchestrator that's running a fan-out (one coach per expanded query). Receives a date window, a single query, and an arXiv category list; fetches all four sources in parallel via the fetch-* skills; dedupes within the four sources for this one query (DOI first, then normalized title + first-author surname); returns the deduped paper-records list as a JSON array directly in the response. Writes no files.

Evaluates Yahoo Fantasy Baseball trade offers. Runs advocate (Acceptor) + critic (Rejecter) variants, computes per-category deltas for both sides across all 10 categories (R/HR/RBI/SB/OBP, K/ERA/WHIP/QS/SV), factors in regression, positional scarcity, and playoff schedule. Verdict is ACCEPT, COUNTER (with specific counter), or REJECT. Use when a trade offer arrives, evaluating potential trade, or assessing a 2-for-1 or consolidation move.

Scans Yahoo Fantasy Baseball free agents for weekly waiver claims (FAAB league, $100 season budget). Fires in advocate (Buy) and critic (Pass) variants, synthesizes, and produces ranked ADD + BID $X recommendations with drop candidates. Use for weekly waiver priority review, FAAB bid sizing, prospect call-ups, closer-committee speculation, or injury replacement.

Per-paper structured-extraction worker for ONE paper at ONE pass depth. Spawned by an orchestrator that's running per-paper fan-out (one extractor invocation per paper). Receives a single paper record, an output directory, a week tag, and the pass to run (1, 2, or 3). Reads the paper at the requested depth (abstract for Pass 1; full PDF via Bash + curl + Read for Pass 2 and Pass 3), invokes the methodology skills in the right order, and writes the combined output to a markdown file. Pass 1 creat

Reverse-engineers a product from the outside to produce a layered strategist analysis of its vision, competitive strategy, tactical initiatives, operational surface, and the ML and systems architecture likely sitting behind its key features. Captures freeform per-step reasoning as analyst working notes in a scratchpad directory, consolidates those notes into a structured analyst-style markdown report following the strategist-voice house style, and renders that report to PDF via the markdown-to-p

An elite forecasting agent that orchestrates reference class forecasting, Fermi decomposition, Bayesian updating, premortems, and bias checking. Adheres to the "Outside View First" principle and generates granular, calibrated probabilities with comprehensive analysis. Use when user asks for forecast, prediction, or probability estimate.

Weekly ML/systems signal radar for substacker. Scans ~30-50 curated intuition-first sources (Olah, Weng, Karpathy, Jay Alammar, Raschka, Willison, Transformer Circuits, Interconnects, arXiv cs.LG, Hugging Face papers, etc.) for signal-not-noise items, cross-references against topic-ledger, produces a lean digest of ≤10 items. Weekly Friday evening. Not daily. Trigger keywords: trends, ML news, weekly digest, signal, watchlist, external signal, what's new in ML.

Times the five single-use chips (Wildcard, 12th Man, Maximum Captain, Clean Sheet Shield, Qualification Booster) across the whole FIFA World Cup Fantasy tournament horizon. Maintains a living deployment plan — which chip is earmarked for which upcoming round, the trigger condition that fires it, and the fallback — and answers the per-round question "fire a chip this round, or hold?". At the verify stage it red-teams the evolution engine's offspring on whether a chip actually belongs on a candida

The recombination core of the FIFA World Cup Fantasy system. Takes the population of candidate squads/plans produced by the parallel archetype strategists and runs the genetic operators from the Evolution document — scores risk-adjusted, rank-relative fitness; selects an elite set spanning the variance spectrum; recombines elites at BUILDING-BLOCK boundaries (captain core, clean-sheet spine, differential pod…) with a feasibility repair operator that never guts a working block; mutates; and enfor

The tournament-theory pillar of the FIFA World Cup Fantasy backroom. Rates per-team fixture difficulty for the round (and across the horizon), estimates progression probabilities (group qualification + each knockout round) via reference-class base rates updated Bayesianly on form, flags weak-group mismatches (powerhouse vs minnow -> attacker captain candidates + defender clean-sheet locks) and fixture SWINGS to plan transfers around. Emits a `fixture` signal that feeds wc-fitness-eval's progress

Optimises the in-round levers for FIFA World Cup Fantasy — the strongest formation-valid XI from the 15, the CAPTAIN LADDER (ordered armband choices across the round's staggered kickoffs with bank-or-switch thresholds), bench order by kickoff time, and manual-sub triggers. This is the single biggest realisable edge in the game (~20–40 pts/round over a set-and-forget manager). Runs in the VERIFY stage of the evolution loop to adversarially review the offspring's matchday plans (emitting keep/anno

Estimates the field for FIFA World Cup Fantasy — public ownership and EFFECTIVE ownership (EO = field_own% × (1 + captaincy_share)), the template set vs the differential list, where the field is OVER-CONCENTRATED (a fade opportunity) and where it is UNDER-REACTING to fixture/rotation news (a get-there-first edge), plus coarse tracking of the visible mini-league rivals for leverage moves relative to specific opponents rather than the global field. Emits an `ownership` signal that feeds wc-fitness

Player-minutes/role/fitness data spine for FIFA World Cup Fantasy — the live web-search engine the population depends on. For a given player pool and round, web-searches the load-bearing facts (predicted XIs that firm ~24h pre-kickoff, injuries/knocks, suspensions including yellow-card accumulation before knockouts, set-piece & penalty duty, recent minutes, form/xG-xA) and turns them into the per-player EV inputs wc-player-ev consumes (start_prob, p60, npxg90, xa90, setpiece_share, pen_taker, ro

Constraint-and-value specialist for FIFA World Cup Fantasy. In the VERIFY stage, red-teams the evolution engine's recombined offspring for BB6 feasibility (budget, nation cap, valid 2-5-5-3 with >=1 legal XI) and value-per-million, emitting a keep/annotate/kill `verify` verdict. During transfer windows and squad builds, generates concrete transfer plans — evaluating transfer ROI against the free-transfer budget and any points hit, handling elimination-driven forced swaps, and scoping the group->

A single population member in the FIFA World Cup Fantasy evolution engine. Spawned once per archetype (genotype) and run in parallel with its siblings by wc-director. Develops its assigned archetype philosophy into ONE complete, valid candidate — a full 15-man squad, a full matchday plan (XI/captain ladder/bench/subs), or a transfer plan — by weighting the shared scout/EV/fixture/ownership signals according to its genotype, then self-stress-tests it with an internal advocate/critic pass before e

Audio cross-correlation alignment for N recordings of the same event. Outputs `.sync.json` sidecars per file — never re-encodes the originals.

Grow X followers via numbered, A/B-testable growth experiments — each with a hypothesis, target metric, rollback state, and verdict. North-star = new-follower ÷ profile-visit conversion.

OpenTrace knowledge graph — search, traverse, and analyze indexed codebases

Outcome-focused Firecrawl workflow skills for research, SEO audits, QA, knowledge bases, lead lists, dashboard reporting, shopping, and design-system extraction.

Use when the user needs academic paper retrieval — searching scientific literature by author/year/journal, finding paper chunks for RAG-style citations, or expanding original text around a known paper offset. Provides six Sciverse tools (search_papers, semantic_search, list_catalog, list_paper_relations, read_content, get_resource) via the sciverse-mcp-server MCP server.

Survey existing contracts and policy documents

Audit AI spend — per-model cost breakdown, top consumers, waste identification, optimization levers.

Map AI cost topology — billing attribution, team-level spend, forecast vs actuals, alert gaps.

Legal Researcher — case law synthesis, statute analysis, jurisdiction comparison

Identify open legal questions and research gaps

Deep clause-by-clause analysis with risk scores

Scan existing contracts for risk patterns

Evaluate model performance — check for accuracy drops, data drift, and error patterns. Use when asked about "model accuracy dropped", "evaluate the model", "check for drift", or "model performance".

Build an ML pipeline — from data to trained model to serving endpoint. Use when asked to "build ML model", "train a model", "prediction pipeline", "classification", or "regression".

ML reconnaissance — inventory all models, pipelines, data sources, and monitoring. Use when asked "what ML do we have", "model inventory", or "ML assessment".

Competitive analysis ending in a clear positioning call — where to play, how to win. Use when asked to "analyze competitors", "competitive landscape", "how do we compare to X", "competitive positioning", "where should we play", "find our white space", or "who else does this".

Audit current model deployment topology — serving config, latency profile, version inventory.

Feedback synthesis — cluster support tickets, NPS verbatims, app store reviews, and churn surveys by theme, separate signal from noise, and produce an actionable insight report. Use when asked to "synthesize this feedback", "analyze support tickets", "what are users complaining about", "NPS analysis", "churn feedback synthesis", or "what's the feedback telling us".

Run a user interview — produce an interview guide and synthesize the output into an actionable insight report. Use when asked to "run a user interview", "synthesize these interview notes", "what do users actually want", "build a persona from this feedback", "find the JTBD in these transcripts", or "analyze this interview data".

Jobs-to-Be-Done analysis — given a product, user descriptions, transcripts, or tickets, produce a JTBD job map with switching forces analysis and opportunity ranking. Use when asked to "find the JTBD", "what jobs are users hiring us for", "job mapping", "what are users really trying to do", "JTBD framework", or "why are users switching".

User segmentation and persona creation from mixed data sources — analytics, CRM, support tickets, reviews, or any combination. Use when asked to "build personas", "who are our users", "segment our users", "create user profiles", "define user archetypes", or "who is the target user".