Productivity & Automation for Claude Code

Agent skills for automating WhatsApp interactions and workflows. Enables sending messages, managing conversations, and automating messaging-based processes.

Surgical, non-interactive git hunk control for AI agents

Master orchestrator for the conference-scheduling pipeline. Runs the four-stage state machine (ingest -> cluster -> elicit -> schedule) across a team of specialist agents, talking to them only through schema-shaped artifacts, verifying each stage's confidence at a gate before advancing, holding and integrating outputs rather than passing them through, freezing its own safety logic against an invariants lock, enforcing the elicitor's outlier-probe diversity floor, surfacing conflicts to the human

Stage 3 collaborative preference agent for the conference-scheduling pipeline — the one a person actually talks to. Reads the theme map and representative talks first so it is grounded, then asks a SMALL number of sharp, choice-based questions contrasting real sessions, holds the person's interests as an uncertainty region it narrows with each answer, deliberately probes outliers to fight selection bias, and elicits the user-owned objective weights and hard constraints. Stops as soon as the sche

Stage 1 ingestion worker for the conference-scheduling pipeline. Parses a conference program (any format) into normalized event records with per-field confidence and independent axes, flags thin abstracts, fans out the high-volume web enrichment to conf-enrichment-researcher sub-workers, merges the enrichment back, and writes a single events.json plus a human index. Reusable across conferences — takes the source and output paths as explicit inputs and hardcodes no conference specifics. Returns t

Stage 4 scheduling worker for the conference-scheduling pipeline. Turns the event records, theme affinities, and preference profile into a feasible day-by-day schedule by solving a constraint-optimization problem — hard constraints (no overlap, travel time, capacity, the attendee's must-attends/blackouts) plus the user-owned weighted objective (interest, breadth, pacing/contiguous-free-time, serendipity). Surfaces unbreakable conflicts as decisions instead of silently picking, and names what the

Turns each published substacker essay into platform-native rewrites. Primary platforms: LinkedIn post + Substack Note + cross-poster blurb (writer's preferred distribution surface). Optional: X thread (generated only if the essay translates well to X; otherwise skipped cleanly). Platform-native reshaping, not paste-same-text-everywhere. Preserves the writer's voice. User posts manually — no auto-posting. Use within 24h of any published post or on specific-platform re-translation requests. Trigge

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

Continuous vigilance specialist for a household finance pipeline. Watches for missed bills, late recurring payments, duplicate charges within 48 hours, statistical anomalies (>3σ above merchant historical average), first-time large merchants, unexpected fees, and identity-theft signals. Produces severity-tagged alert files with transaction ids, evidence, and recommended actions (call bank, freeze card, dispute, monitor, negotiate fee). Use after every drop, when checking for anomalies post-state

Meticulous bookkeeper for a household finance pipeline. Takes a manifest produced by the intake classifier and the underlying PDFs, extracts every transaction or holding, normalizes amounts to integer cents and dates to ISO 8601, deduplicates against the existing transaction store, categorizes via taxonomy rules + LLM fallback, reconciles statement math, and writes append-only updates to transactions.json, balances.json, investments.json, retirement.json, hsa.json, and mortgage.json. Halts and f

Master orchestrator and synthesizer for the household finance team. Runs the per-drop pipeline (intake → bookkeeper → spending/vigilance/savings/investments/tax in parallel → CFO synthesis), the monthly briefing pipeline, the weekly dashboard generation, and the on-demand chat mode where the user asks ad-hoc finance questions. Produces the one-page monthly briefing the household actually reads — net-worth snapshot, four numbers (income / spend / savings rate / Δ net worth), three wins, three iss

Document intake specialist for household finance pipelines. Classifies every PDF in a drop folder by document type (checking, savings, credit card, brokerage, 401k, HSA, mortgage, tax form, insurance), matches each against existing accounts by institution + mask, proposes new account records when no match exists, archives PDFs by statement period (not drop date), and produces a manifest.json for downstream agents. Use as the first stage of any per-drop pipeline, when adding a new batch of bank o

Manages the household short-to-medium-horizon balance sheet — emergency fund, sinking funds, mortgage, any consumer debt, and credit-card rewards optimization. Tracks goal progress and pacing, sizes the emergency fund against essential expenses, ranks debts by APR for avalanche payoff, models mortgage prepayment vs market return, identifies bills due for renegotiation, and surfaces every recommendation with the dollar impact attached. Use as the savings-debt phase of the monthly briefing, when r

Tracks tax-relevant events year-round so April is a non-event. Maintains the expected-document checklist (W-2, 1099-DIV/INT/B, 1098, 5498-SA, 1095), accumulates deduction candidates from transactions, manages the HSA receipt vault for deferred reimbursement, surfaces tax-aware moves throughout the year (TLH, Roth conversion windows, charitable bunching, backdoor/mega-backdoor Roth eligibility), validates estimated tax payments against safe-harbor rules, and assembles a year-end packet for the ho

Julian project tools — deploy, manage, and interact with Julian instances

Ingests and indexes the writer's Substack corpus. Watches the substacker inbox/ for new material (markdown notes, transcripts, Claude conversation exports, Readwise highlights), normalizes format, tags by topic and intuition density, dedupes against existing corpus, maintains the topic-ledger, and sweeps stale seeds. Use at session start, when the user drops raw material into inbox/, when asking "what have I written about X", or when user mentions ingest, library, corpus, seeds, indexing, topic

Orchestrates a multi-agent team for Yahoo Fantasy Baseball management. Spawns specialists (lineup, waiver, streaming, trade, category, playoff) each in advocate + critic variants, runs dialectical-mapping-steelmanning synthesis, deliberation-debate-red-teaming stress tests, and produces plain-English morning briefs for a user with zero baseball knowledge. Use when running morning brief, weekly kickoff, evaluating trades, or any Yahoo Fantasy Baseball decision for the user's league.

Picks today's Yahoo Fantasy Baseball lineup for a 26-roster H2H Categories league with OBP/QS scoring. Fires in two variants — advocate (steelmans each start) and critic (red-teams each start) — then synthesizes per slot. Emits daily_quality signals and writes a START/SIT decision per active roster position. Use for daily lineup optimization, start/sit calls, platoon decisions, or daily Yahoo Fantasy Baseball management.

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.

Orchestrates the end-to-end transformation of a methodology document (PDF, markdown, book chapter, research paper, framework guide) into a properly structured Claude Code skill - SKILL.md plus optional resource files plus an evaluation rubric. Runs Adler's six-step reading-and-construction methodology - inspectional reading, structural analysis, component extraction, synthesis-application, skill construction, validation - by sequentially invoking the standalone skills that own each step. Each st

Vibes is for people who use Claude Code, but don't know how to code.

User-facing orchestrator for FIFA World Cup Fantasy. Runs the evolutionary generation loop — spawns archetype strategist teams in parallel, drives fitness/selection/recombination via wc-evolution-engine, has specialists adversarially verify the offspring, then presents a DECISION BOARD of 2-4 weighted options and STOPS for the expert manager to choose. Advisory not prescriptive: surfaces options with full reasoning and dissent, never auto-commits a squad/XI/captain/transfer/chip. Manages the tou

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

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

Slack command that embeds healthy skepticism into AI dialogue by countering confirmation bias and authority bias. Encourages critical evaluation of AI responses.

Drop-in agentic loop skills (autoresearch, literature & writing, data, optimization, red-team, power analysis). Skills install namespaced as agent-loops:<name>.

Claude Essentials - Complete development toolkit with 1 command, 18 skills, and 4 agents unified under ce namespace

Enforce measurement-driven optimization and reproduce-first debugging. Activates automatically on performance and bug-related queries.

MCP server plugin for voice input and output in Claude Code, enabling voice-based interaction with coding assistants.

Claude Code guide — interactive onboarding walkthrough and Q&A on setup, best practices, automation, and effective workflows

Comfy Cloud in Claude Code: the hosted MCP connection plus slash commands for generating images, video, audio, and 3D and searching the catalog.

Efficient skill management with progressive discovery — 410+ production-ready skills across 33+ domains.

Prompt engineering skill for Claude Code, OpenClaw, Codex, Grok CLI, and Hermes Agent

Holistic state-of-the-project audit: unmerged branches, stalled PRs, failed GitHub Actions, stale builds, plan drift, unreleased commits, log errors. Read-only investigation → grouped checklist → fix only after confirmation.

Final-composite step: burn SRT into pixels via libass, mix in a dub track, keep the original audio as a low-volume bed — all in one ffmpeg encode (no cascade).

Delete 引流 / emoji spam replies (同城面付 / 寻固炮 / 线下上门 etc.) under your X posts. Covers the last 7 days via X recent-search.

Turn a 王建硕-style WeChat article into a narrated 1080×1920 portrait MP4 (30–90s) — Volcano TTS voiceover, HyperFrames CSS/GSAP per-scene animation, subtle SFX, abstract watercolor background.

Capture a writer's voice from a few sample articles and rewrite text to sound like that author. Works for any author, 王建硕 included.

Time-aligned TTS dub of a video against a target-language SRT. Volcano TTS for Chinese, edge-tts for everything else. Single-speaker by default; opt-in visual diarization for multi-speaker.