OneWave AI Claude Skills

Skill

200+ single-file skills, mostly for sales, marketing, consulting and operations: deal strategy, forecasts, outreach emails, proposals and SOWs, contract review, resume screening, expense checks and RFP responses, plus personal finance and job search. Contract and tax output needs expert review.

OneWave-AI/claude-skills348
Открыть на GitHub
README

Claude Skills Library

Production-ready skills for Claude Code. Built and maintained by OneWave AI.

229 skills across three pillars: business (sales, marketing, consulting, ops), everyday life (personal finance, travel, fitness, job hunting), and coding (engineering, design, AI agent architecture).

Companion repo: Open Agent Stack -- 27 artifacts that need more than a single file. Highlights: vibe-stack (14 skills: idea to shipped app, with a real motion system and an AI-slop detector), sales-desk (the rep's whole daily loop), and 7 design-token themes. Skills here stay single-file and zero-dependency; anything with a manifest, a team, or a build step lives there.


Deploy 3 to 50+ independent Claude agents in parallel, each with its own 1M token context window. Each agent spawns sub-agents underneath. A top-tier commander (Fable or Opus) orchestrates the swarms, and you pick a power level -- Max Power, Heavy, Balanced, or Economy -- that sets the Opus/Sonnet/Haiku model mix for each layer. The system runs in waves -- execute, audit, propagate -- and checks its own work between each.

Built for tasks where one agent isn't enough: large refactors, full-site audits, bulk content generation, codebase migrations. Battle-tested on 60+ concurrent agents in a single session.

View agent-army


Getting Started

# Claude Code: install a single skill
claude skill install OneWave-AI/claude-skills/<skill-name>

# Or clone the full library
git clone https://github.com/OneWave-AI/claude-skills.git ~/.claude/skills

# OpenAI Codex CLI (same files, unchanged)
git clone https://github.com/OneWave-AI/claude-skills.git ~/.codex/skills

Each skill is a self-contained SKILL.md file. No dependencies, no build step.

These skills follow the Agent Skills open standard, so the same files work unchanged in Claude Code, OpenAI Codex CLI, Gemini CLI, Cursor, and 30+ other tools -- point your tool's skills directory at the clone and go.


Categories

AI Agent Architecture

Skills for building, orchestrating, and managing autonomous AI agents.

SkillDescription
agent-army2-layer parallel agent hierarchy for large tasks at maximum speed
agent-to-agentA2A communication protocol -- message passing, shared context, handoffs
agent-swarm-deployerDeploy swarms of sub-agents for massive parallel data processing
agent-team-builderDesign and deploy custom agent teams for business workflows
sub-agent-orchestratorParent/child agent orchestration with task delegation
scoutRecommends the best skill for any task
scout-proEnhanced scout with skill chains, pattern recognition, usage learning
skill-navigatorGuide to all skills with combination recommendations
skill-composer-studioChain multiple skills into custom workflows
mcp-server-builderBuild, wire, and debug an MCP server that connects Claude to your systems
jev-integrateWire a System One model (Jev/Von) into a feature -- typed decisions, gates, thresholds
jev-evalSweep criteria wordings and thresholds for a typed-decision config, with a runnable harness
jev-auditFind LLM calls that are really classifications, and price the swap

Agent Workflow and Context Engineering

How to actually run a coding agent: what to put in context, what to write to disk, and the guardrails that keep the output honest.

SkillDescription
plan-in-filesPlans that live on disk and survive a context reset, crash, or handoff
session-handoffCompress a session into a handoff doc the next session can resume from
context-budgetSpend the context window deliberately -- read narrow, silence noise, compact on time
ai-coding-guardrailsThe ten rules that catch an agent's known failure modes before the diff lands
grill-the-briefInterrogate a vague request before writing code, with defaults so nothing stalls
claude-md-writerWrite the CLAUDE.md / AGENTS.md that teaches an agent your repo's commands and traps
fresh-library-docsCheck the installed version and real signatures before calling an API from memory
agent-eval-harnessA 20-case eval set that catches prompt, model, and tool regressions before customers do

Anthropic / Claude Products

Skills built around specific Anthropic product releases.

SkillDescription
overnight-repo-auditorUses Managed Agents (14.5hr runtime) for autonomous overnight codebase audits
multi-agent-client-onboardingAgent SDK: 3 parallel agents for client assessment
gmail-to-crm-pipelineMCP Connectors: Gmail to CRM lead qualification pipeline
full-codebase-migrator1M context window: ingest entire codebases for migration planning
claude-design-system-architectGenerate a premium design system (tokens, type, motion) exported to Tailwind/CSS
claude-landing-composerBuild premium animated landing pages in Next.js + Framer Motion, anti-template
claude-design-criticAudit a UI and de-AI it — design + copy fixes toward editorial/premium

Claude Cowork

Folder-first knowledge-worker workflows built for Claude Cowork -- now with cloud sessions, scheduled tasks that run with no device online, and Microsoft 365 write tools (July 2026).

SkillDescription
cowork-deal-roomMulti-step deal room due diligence -- contracts, financials, risk scoring
cowork-folder-organizerTurn a messy folder or shared drive into an organized, indexed workspace with an undo trail
cowork-inbox-triageDaily inbox sweep -- classify, draft replies in your voice, queue follow-ups; scheduled-task ready
cowork-hiring-screenerScore a folder of resumes against the JD -- ranked shortlist with evidence, interview kits, drafts
cowork-expense-auditReconcile receipts to statements, categorize, flag policy violations, output the expense report
cowork-rfp-responseShred an RFP into a compliance matrix, mine your past proposals, draft the response honestly
cowork-invoice-chaserAR aging report plus escalating dunning drafts matched to invoice age; weekly scheduled task
cowork-contract-renewal-radarExtract every auto-renew clause and notice window, calendar the real decision deadlines
cowork-data-room-builderPrep YOUR data room for a raise or acquisition -- checklist, gap report, organized structure
cowork-vendor-comparisonNormalize vendor quotes into one matrix -- true TCO, buried terms, negotiation leverage
cowork-qbr-builderClient folder -> QBR with value receipts, honest misses, health read, expansion asks
cowork-sop-writerTurn a process walkthrough into a clean SOP -- steps, decision points, tribal knowledge captured
cowork-calendar-defragMeasure meeting load and fragmentation, propose cuts and focus blocks, draft the decline messages
cowork-tax-prep-organizerSweep folders for tax docs, build the CPA-ready package and the missing-form chase list
cowork-medical-bill-auditorMatch bills to EOBs, catch overbilling and unsubmitted claims, draft dispute letters
cowork-home-inventoryPhotos + receipts -> insurance-grade home inventory with values, serials, and the gap list

Sales and Revenue

SkillDescription
deal-closer-playbookClosing strategy with buying committee mapping and objection handling
renewal-predictorPredict renewal likelihood from health score signals
expansion-revenue-finderIdentify upsell and cross-sell opportunities in existing accounts
pipeline-health-analyzerIdentify stalled deals, predict close probability
deal-review-frameworkMEDDIC/BANT deal assessment with risk scoring
deal-momentum-analyzerScore deal velocity from engagement patterns
sales-forecast-builderWeighted pipeline forecast with scenario modeling
sales-call-prep-assistantPre-call research briefs with discovery questions
sales-methodology-implementerMEDDIC, BANT, Sandler, Challenger, SPIN implementation
lead-scoring-modelBuild custom lead scoring from historical win/loss data
inbound-lead-qualifierScore inbound leads by ICP fit, intent, and urgency
cold-email-sequence-generatorMulti-touch outbound campaigns optimized for response
personalization-at-scalePersonalized first lines for hundreds of prospects
champion-identifierFind internal champions in target accounts
intent-signal-aggregatorMonitor buyer intent signals across the web
objection-pattern-detectorMine lost deals for recurring objection patterns
lookalike-customer-finderFind companies matching your best customer profile
quota-setting-calculatorTop-down vs bottom-up quota models
sales-comp-plan-designerVariable compensation design with accelerators
sales-coaching-plan-generatorIndividualized rep development plans
ramping-rep-tracker30/60/90/120 day ramp milestones
rep-performance-scorecardMulti-dimensional rep evaluation
territory-planning-optimizerAccount assignment by revenue potential
icp-deep-scannerRead-only deep scan of connected tools → data-grounded ICP + persona library
customer-panel-of-expertsYour buyer personas debate any decision (launch, price, product) and recommend
prospect-panel-simulatorSimulate prospects to pressure-test emails, decks, and pages before sending
pricing-change-strategistPlan a price increase: segmentation, scenarios, grandfathering, full comms kit

Consulting and Professional Services

SkillDescription
client-proposal-generatorFull consulting proposals from a brief
sow-generatorProfessional Statements of Work with legal boilerplate
client-health-dashboardRAG status across all client accounts
churn-autopsyPost-mortem analysis when a client churns
onboarding-checklistCustomized client onboarding plans
ai-readiness-assessmentAssess how ready a business is for AI adoption
saas-replacement-plannerEvaluate which SaaS tools can be replaced with AI agents
roi-calculatorAI implementation ROI with sensitivity analysis
meeting-intelligenceExtract decisions, action items, and sentiment from transcripts
meeting-to-tasksConvert transcripts to action items with owner assignment
weekly-business-reportAuto-generated weekly KPI reports

Engineering and DevOps

SkillDescription
code-review-proSeverity-ranked security (OWASP Top 10:2025), correctness, and performance review
debug-like-expertMethodical investigation with hypothesis testing
api-load-testerProgressive load testing with bottleneck analysis
database-migratorCross-provider database migration with validation
incident-responderProduction incident response automation
runbook-generatorOperational runbooks from codebase analysis
data-pipeline-builderETL/ELT pipeline design and implementation
dependency-auditorSecurity vulnerabilities and outdated packages
test-coverage-improverFind and fill test coverage gaps
docker-debuggerContainer troubleshooting and optimization
typescript-migratorJavaScript to TypeScript migration
env-setup-wizardEnvironment variable management
error-boundary-creatorReact error boundaries and fallback UIs
git-pr-reviewerPull request quality review
regex-debuggerVisual regex breakdown and debugging
performance-profilerApplication performance profiling
api-endpoint-scaffolderREST API endpoint generation
react-component-generatorReact components with TypeScript and a11y
database-schema-designerQuery-driven schemas, indexes, ERDs, and safe migrations (PostgreSQL 18 default)
landing-page-optimizerConversion and performance optimization
browser-test-loopDrive a real browser to prove a UI change works -- console errors included
static-analysis-sweepSemgrep, secret, and CVE scanning over a diff, triaged into fix-now vs noise

Design Systems and UI

Build, install, and enforce design systems. Pairs with the 7 ready-made design-token themes in Open Agent Stack design-styles/ -- aurora-mesh, cirrus, liquid-glass, mono-brutalist, neo-terminal, sand-terra, tidal.

SkillDescription
design-style-installerInstall a complete token theme (from Open Agent Stack or local tokens.json) and repaint the project
design-tokens-syncFind and fix token drift -- stale copies, hardcoded values, orphans -- plus a CI guard
motion-language-designerDuration/easing scales, choreography rules, signature moves -> Framer Motion + CSS
dark-mode-converterReal dark mode via semantic tokens and elevation logic, not naive inversion
typography-scale-builderFluid type scale with per-step line-height/tracking, roles, vertical rhythm
design-system-generatorDesign tokens, components, documentation
css-animation-creatorProfessional animations and micro-interactions
responsive-layout-builderCSS Grid, Flexbox, container queries
screenshot-to-codeScreenshots to React 19 + Tailwind v4 code, with a render-and-compare loop
color-palette-extractorExtract palettes from images or sites -- HEX/RGB/Tailwind/CSS variables
font-pairing-suggesterFont pairings with hierarchy and loading strategy
brand-consistency-checkerScan documents and slides for off-brand colors, fonts, logos
accessibility-auditorWCAG compliance audit and fixes

Security and Compliance

SkillDescription
compliance-checkerGDPR, HIPAA, SOC2, CCPA, PCI-DSS audit
security-pentest-plannerPenetration test planning (OWASP Top 10)
tech-due-diligenceTechnical due diligence for M&A/investment
contract-analyzerReview contracts for concerning clauses
contract-redlinerGenerate redline suggestions with replacement language

Marketing and Content

SkillDescription
seo-optimizerKeyword analysis, readability, competitor comparison
seo-keyword-cluster-builderTopic cluster architecture
landing-page-copywriterHigh-converting copy using PAS, AIDA, StoryBrand
brand-voice-analyzerExtract and codify brand voice from existing content
content-repurposerTransform content into 8+ formats
social-repurposerAdapt content for different platforms
social-selling-content-generatorLinkedIn thought leadership posts
linkedin-post-optimizerProfessional narrative with hooks
utm-parameter-generatorStandardized UTM tracking
competitor-content-analyzerTrack competitor content strategy
competitor-price-trackerMonitor competitor pricing changes
competitor-intel-agentComprehensive competitor monitoring
customer-review-aggregatorAggregate and analyze reviews from G2, Capterra, etc.
podcast-content-suiteTransform podcasts into content marketing
webinar-content-repurposerWebinar to blog, social, email
email-template-generatorProfessional email templates
email-subject-line-optimizerA/B test subject lines
product-launch-war-roomAdversarial GTM war room: go/no-go, risk register, phased rollout, kill criteria
hyperframes-ad-directorBrief → finished HyperFrames video ad: hook, script, storyboard, scenes, cuts
hyperframes-sales-demo-builderPersonalized product-demo videos in HyperFrames for a specific account
hyperframes-explainer-builderURL or docs → tight 60-90s product explainer with walkthrough scenes and VO
hyperframes-testimonial-builderReal reviews and case-study results → social-proof videos: spotlights, review walls, proof reels
hyperframes-local-promo15-30s local-business promos: menu drops, offers, events, hiring clips

Strategy and Finance

SkillDescription
pricing-strategyPricing model design with competitive analysis
market-sizingTAM/SAM/SOM with top-down and bottom-up estimates
pitch-deck-reviewerInvestor deck review with scoring
board-deck-generatorBoard meeting presentation content
investor-update-writerMonthly/quarterly investor updates
executive-dashboard-generatorData to executive-ready reports
financial-parserExtract data from invoices, receipts, statements
portfolio-analyzerInvestment portfolio risk and diversification
budget-optimizerSpending analysis and savings strategies
financial-goal-plannerSavings targets and investment strategies
tax-strategy-optimizerPre-tax, Roth, charitable giving optimization

Operations and People

SkillDescription
workflow-automatorDesign automated workflows from manual processes
okr-generatorOKRs following Google/Intel methodology
customer-journey-mapperFull journey from first touch to advocacy
hiring-scorecardStructured scorecards for any role
knowledge-base-builderFAQ identification and tutorial creation
technical-writerUser guides, architecture docs, onboarding materials
doc-coauthoringStructured documentation co-authoring workflow
job-application-optimizerTailor resumes to job postings
raise-negotiation-prepSalary research and negotiation scripts

Work Tools

Skills that ship scripts and reference data, not just instructions -- the jobs a model gets wrong on its own.

SkillDescription
bookkeeping-closeCategorize, reconcile bank to ledger, and run month-end close from QuickBooks/Xero/bank exports -- never plugs a difference
spreadsheet-model-auditorScript audit of any xlsx model: hardcodes, broken ranges, circular refs, inconsistent formulas, totals that do not foot
bi-measure-builderDAX, Tableau calcs and LookML that total correctly -- filter context, LODs, time intelligence, with a checker script
trade-quote-builderEstimates and bids for trades and field services -- waste, burden, overhead, margin vs markup, client xlsx/PDF
crm-data-cleanupVendor-neutral CRM dedupe and normalization -- confidence tiers, a dry-run merge plan, shared-inbox traps caught
devils-advocateThe strongest honest case against a decision you are leaning toward -- steelman, pre-mortem, no reflexive contrarianism
statement-extract-and-proveBank, card and brokerage statement PDFs to CSV/XLSX, proven: opening + transactions = closing, first broken row pinpointed
spreadsheet-qaAnswer questions about your own exports correctly -- profile, data dictionary, compute in code, show the rows and filters used
template-deck-builderDecks built into your company's slide master and placeholders, storyline first, rendered and QA'd for overflow and off-brand
sec-filing-puller10-K/10-Q numbers from SEC EDGAR XBRL with a citation per number -- fiscal alignment, restatements, derived Q4, TTM

Small Business Essentials

The problems that actually kill small businesses -- cash, margins, visibility, people, the lease.

SkillDescription
cash-flow-forecaster13-week rolling cash forecast from bank exports and AR/AP -- crunch weeks circled with the levers to pull
job-profitability-analyzerWhich clients and jobs actually make money -- loaded labor costs, effective hourly rate per client
local-seo-optimizerWin the Google local pack -- Business Profile audit, review engine, NAP sweep, location pages
review-response-writerRespond to every review in the owner's voice -- gracious 5-stars, masterful 1-stars, escalation flags
employee-handbook-builderPlain-English handbook that matches how the business actually runs, with state-law check flags
inventory-reorder-plannerSales history -> reorder points, this week's purchase list, and the dead-stock cash report
job-post-writerJob posts the right person actually applies to -- pruned requirements, honest sell, pay handled
commercial-lease-analyzerTrue all-in occupancy cost, the small-tenant trap clauses, and the negotiation ask list

Skill Format

Every skill is a single SKILL.md file with YAML frontmatter:

---
name: skill-name                 # required — kebab-case, matches the folder
description: What the skill does and when to use it.   # required — drives auto-selection
tools: Read, Write, Bash, Agent  # optional — restrict tool access; omit to inherit all
model: inherit                   # optional — pin a model; omit to inherit the session model
---

# Skill prompt content here...

Claude Code loads this as a system prompt when the skill is invoked. Only name and description are required; most skills here use just those two. The description is what Claude reads to decide when to trigger the skill, so make it specific.


About OneWave AI

OneWave AI is a boutique AI consulting firm based in Florida, specializing in Claude and the Anthropic ecosystem. We help businesses from SMB to enterprise implement AI that ships real results -- from Claude for Enterprise deployment to custom agent architecture.


Contributing

  1. Fork the repository
  2. Create a new folder with your skill name
  3. Add a SKILL.md following the format above
  4. Submit a pull request

Skills should be production-ready, well-documented, and solve a real problem. No placeholder or stub skills.


License

MIT