desktop
Local-first AI agent that works across your files, apps, and browser
All Skills in This Repository (131)
Pure Emerald Level Indicatorscustomer-escalation
Package customer escalations into structured briefs for engineering, product, or leadership.
kb-article
Draft a publish-ready knowledge base article from a resolved support issue.
customer-research
Synthesizes multi-source customer research into structured briefs with attribution and confidence scoring.
draft-response
Draft tailored customer-facing responses for support tickets, emails, and live chats.
user-research
Plan, conduct, and synthesize user research studies into actionable insights.
research-synthesis
Synthesize user research data into themes, insights, and prioritized recommendations.
ux-copy
Generate and review UX copy for web and mobile interfaces.
design-handoff
Generate structured markdown developer handoff specs from Figma URLs or design descriptions.
design-critique
Provide structured design feedback on usability, visual hierarchy, and accessibility.
accessibility-review
Audit designs or pages for WCAG 2.1 AA accessibility compliance.
design-system
Audit design systems for naming consistency and hardcoded values.
search-strategy
Decompose natural language questions into parallel source-specific searches across connected data stores.
Frequently Asked Questions
FAQPage SchemaHow to install OpenYak?▼
Run `npx skills add openyak/desktop --all -g -y` in your terminal to install all skills in this suite globally.
What can OpenYak do on my computer?▼
It reads your local files (DOCX, XLSX, PDFs), works in native apps and a managed browser, and produces briefs, analyses, and plans while keeping all data on your device.
Does OpenYak require a cloud account?▼
No. It runs local-first with no login required, and you can use local models via Ollama or bring your own API keys for OpenAI, Anthropic, and other providers.
What skill packs are included?▼
The suite ships plugins for sales outreach and forecasting, marketing campaigns and SEO, legal contract review, finance close and reconciliation, HR and recruiting, data analysis and SQL, design critique, product specs, and bio-research pipelines.
Can I control the agent while it works?▼
Yes. You can watch the live Computer or Browser workspace, take over for any step yourself, then hand control back to the agent without restarting the task.
Related Repositories in Personal & Entrepreneur
View All in Personal & Entrepreneur→clawdis
Run a personal AI assistant across your devices and chats
awesome-claude-skills
Curated library of 1000+ ready-to-use AI agent skills
chatgpt-on-wechat
Personal AI assistant that plans tasks, remembers, and chats anywhere