65 verified Runes · install in one command

Your AI stack,
already assembled.

Pick a workflow. Run one command. Everything installs, connects, and executes — automatically.

Deterministic runs. 70% fewer tokens. Weeks of setup → one command.

Browse Runes — FreeHow it works

THE CORE CONCEPT

What is a Rune?

A Rune is a composable AI workflow — a recipe that chains multiple Skills into one reliable pipeline. Install once, get the same result every time.

Rune Orb
SkillThe Ingredient

A packaged service capability — gmail, slack, github. Each Skill bundles every action for that service with agent-ready instructions.

RuneThe Recipe

A multi-Skill workflow — chained in sequence or parallel, verified with a Trust Score before you run it.

TrustThe Guarantee

Every Rune carries a Trust Score — showing which services it accesses, what scopes it needs, and the risk level of each action.

WHY IT MATTERS

Vibe coding hits a ceiling.
Runes break through it.

92% of devs use AI daily. The edge isn't the model — it's the playbook. Runes make every playbook deterministic, auditable, and cheap to run.

💡

The real cost of AI automation isn't the API bill — it's the 2–4 weeks of engineering to wire a reliable workflow, plus ongoing maintenance when it breaks. Runes eliminate that cost entirely. The token savings are just a bonus.

Reproducible by Design

100% deterministic

LLMs improvise differently every run. A Rune locks the workflow graph — same Skills, same order, same result. Guaranteed.

BEFOREPrompt → LLM freestyle → unpredictable
AFTERRune → verified pipeline → guaranteed result

Up to 70% Fewer Tokens

~12,000 → ~3,600 tokens/run
① Zero-token planning−2,000 tok
The graph is the plan. No LLM call to decide what to do next.
② Context isolation−5,000 tok
Each step gets only its own inputs — not the full session history.
③ Semantic cachingvariable
Matching inputs reuse cached results. LLM skipped entirely.
④ Smart routing−30–50%
Haiku for classification, Groq for extraction, Sonnet only for synthesis.
SkillsBench: Haiku + Rune 27.7% pass rate vs Opus alone 22.0% — cheaper model, better result.
Beyond runtime: Building a reliable workflow from scratch takes 2–4 weeks. Factor in the engineering hours + debugging cost, and Runes save 10–50× more than the token bill alone.

Trust, Don't Hope

Least-privilege scopes

Least-privilege by default. Secrets injected at runtime — never exposed to the LLM. Every execution produces an immutable audit log.

ClawHavoc (Jan 2026): 341 malicious Skills → 9,000+ installs compromised. Snyk: 47% of unvetted Skills had security issues. RuneHub's Trust Score was built for exactly this.
BEFOREUnvetted registry — 47% have security issues
AFTERTrust Score 94 — scoped, audited, verified
65
Runes
139
Skills
228
Actions
68%
Avg Token Savings
87
Avg Trust Score
95%
Skip Manual Setup

HOW IT WORKS

One command. That's it.

RuneHub resolves dependencies, configures auth, wires every step, and runs — all from a single CLI command.

terminal
$ rune install morning-brief --run

Resolving dependencies...
  + gmail          
  + calendar       
  + github         
  + brave-news     
  + openweather    
  + market-pulse   
  + llm-triage     
  + llm-compose    
  + slack          
  + telegram       
  + notion         
✓ 11 skills auto-installed  |  Trust Score: 91  |  44 actions ready

Running morning-brief...
  [1/10] fetch-calendar     ✓  0.2s
  [2/10] fetch-emails       ✓  0.3s
  [3/10] fetch-github       ✓  0.5s
  [4/10] triage-llm        ✓  1.4s
  [5/10] fetch-news         ✓  0.5s
  [6/10] fetch-weather      ✓  0.3s
  [7/10] fetch-market       ✓  0.4s
  [8/10] compose-llm       ✓  1.8s
  [9/10] broadcast-3ch      ✓  0.3s
  [10/10] archive-notion    ✓  0.1s

✓ Done in 5.8s  ·  4,800 tokens used  (saved 71%)  ·  archived to Notion
AUTO-INSTALL

Skills resolve like npm packages — no hunting for dependencies or versions.

AUTO-CONFIGURE

Auth scopes, action wiring, and agent context set in the correct order, every run.

AUTO-AUDIT

Immutable logs per step. Token usage tracked. Nothing executes outside the declared graph.

Live Example·🌅 Morning Brief
Trust Score 9111 skillsDual LLM · 3-output
INPUT — Calendar · Email · GitHub
LLM 1 — Triage & Score
API — News · Weather · Market
LLM 2 — Compose
OUTPUT — Slack · Telegram · Notion
scheduleemailsdev taskspriority matrixheadlinesweathermarketteam briefpusharchiveINPUT📅Google Calendarschedule + conflictsINPUT📧Gmail InboxVIP + flagged emailsINPUT🐙GitHub IssuesPRs · blockers · dueLLM🎯Claude Triageclassify · dedupe · urgency-score all inputsAPI🔍Brave Newsagenda-aware headlinesAPIOpenWeatherforecast + commute riskAPI📈Market Pulseportfolio snapshotLLM✍️Claude Composesynthesize · personalize · adapt to calendar densityOUTPUT💬Slack Post#morning-standupOUTPUT📱Telegrammobile push + actionsOUTPUT📒Notion Archiveknowledge base log

Three input channels — calendar, email, and GitHub Issues — feed Claude Triage, which classifies, dedupes, and urgency-scores everything into a priority matrix. Claude Compose then weaves that matrix with live headlines, weather forecasts, and market data into a brief personalized to your calendar density — broadcast simultaneously to Slack, Telegram, and Notion for long-term recall.

Featured Runes

View all 65
Morning Brief
ProductivityArchmage
🌅
Compile a hyper-personalized daily briefing from calendar, email, GitHub issues, live news, weather, and market data — triaged by AI, composed into a structured brief, and broadcast to three channels simultaneously.
Trust Score99
inputllmapioutput11 skills
Deep Research
ResearchArtisan
🔬
Conduct multi-source research on a topic using web search, academic papers, and AI synthesis.
Trust Score96
inputapillmoutput6 skills
Blog Forge
ContentSmith
📝
Generate a polished blog post from a topic with SEO optimization and publish it.
Trust Score87
inputapillmoutput5 skills
Code Sentinel
DevOpsSmith
🛡️
Automatically review pull requests for bugs, style issues, and security vulnerabilities.
Trust Score85
inputapillmoutput5 skills
Data Insight
AnalyticsSmith
📊
Analyze spreadsheet or database data and generate visual insights with AI interpretation.
Trust Score83
inputapillmoutput5 skills
Support Sage
Customer SupportSmith
🧙
Auto-classify and draft replies for customer support tickets using AI.
Trust Score83
inputllmoutput5 skills
Social Pulse
MarketingArtisan
📡
Monitor brand mentions and sentiment across Twitter, Reddit, and news sources.
Trust Score92
inputapillmoutput6 skills
Inbox Zero
ProductivitySmith
📨
Automatically triage, categorize, and draft replies for email overload.
Trust Score83
inputllmapioutput5 skills
Alpha Watch
FinanceSmith
🐺
Monitor crypto markets, DeFi protocols, and on-chain activity for alpha opportunities.
Trust Score83
apillmoutput5 skills
Standup Sync
DevOpsSmith
🤝
Auto-generate daily standup updates from GitHub activity, Linear issues, and Slack messages.
Trust Score83
inputllmoutput5 skills
Release Radar
DevOpsApprentice
📡
Monitor GitHub releases for tracked repositories and notify the team with changelogs.
Trust Score74
inputllmoutput4 skills
Vuln Sentinel
SecurityArtisan
🔐
Scan code for vulnerabilities across multiple security tools and generate a unified security report.
Trust Score92
apillmoutput6 skills

DEVELOPER ECOSYSTEM

Build a Rune.
Earn from every run.

Publish your workflow to the registry. Every time someone runs your Rune, usage revenue is generated — automatically tokenized and distributed to you. No ads, no subscriptions — pure usage economics.

🔨
Build onceAuthor a Rune with the Builder or CLI. Auto-Fill wires the gaps. Validate with one command.
📈
Community improves itDevelopers fork, extend, and contribute — each Rune evolves through community iteration.
💰
Revenue is tokenizedUsage fees flow back to creators as on-chain revenue shares. Verifiable, automatic, no middleman.
Start building ›
01
Author a Rune
Drag, wire, validate
02
Publish to registry
Trust Score ≥ 70 required
03
Users run it
Every execution is logged
04
Auto-eval & optimize
Community forks improve the Rune
05
Revenue → tokenized
Creator receives 85% of usage fees

PRICING

Start free. Optimize when ready.

Use your own API keys at no cost. Upgrade for smart routing, caching, and guaranteed savings.

FREE

$0 / forever
Install & run any Rune
Use your own API keys
Community Rune registry
Basic execution logs
Get Started
MOST POPULAR

OPTIMIZER

$29 / month
Everything in Free
Smart model routing
Semantic cache pool
Token usage analytics
Priority support
Start Saving

PRO

$99 / month
Everything in Optimizer
50% savings SLA guaranteed
Full audit logs & export
Team workspaces (10 members)
Custom model routing rules
Go Pro

Enterprise? Contact us for dedicated infrastructure, compliance, and custom SLAs.

CREATOR PROGRAM
Build Runes. Earn from every run.
Every time someone runs your Rune, usage revenue is tokenized and 85% flows back to you — automatically, on-chain, no invoicing.
Start building ›

Browse the Skills Registry

139 skills · 228 actions · 65 Runes

View All Skills ›