Build in public: keeping the OSK v4 AI Assistant robust at 250+ skills

Canonical version: Build in public: keeping the OSK v4 AI Assistant robust at 250+ skills.

Build-in-public update on the AI Assistant Architecture inside OSK v4.

Most AI assistant systems collapse past 20 skills because robustness is bolted on at the end. In OSK v4 it's the foundation. I just crossed 250 skills and the system is steadier than it was at 50.

Eight mechanisms making it work:

• Capability indirection. Every skill declares metadata.capability: <domain>.<subject>.<verb>. Cross-skill references go through capabilities, never names. Rename a skill; nothing breaks. Dependency inversion at the skill layer.
• Runtime path resolution, no hardcoding. Today's biggest refactor. About 120 skills used to embed their own copy of "where does the daily note live" logic. Now they declare metadata.note-types and resolve folders, templates and prefixes from osk-cli at run time. One canonical Setup block. Path drift becomes structurally impossible.
• Spec-layered frontmatter, validator-enforced. osk-meta-skill-health went from 10 checks to 18 yesterday. New ones include spec-layering-top-level, capability-collision (one capability equals one skill, hard error), workflow-without-composes, and claude-code-field-nested. The contract is now machine-verifiable.
• Dispatcher consolidation. The moment a second skill would share (subject, verb) with an existing one, I refactor into a polymorphic skill with a --mode flag. osk-action-archive absorbed four siblings this way. SRP at the capability level, not the name level.
• Self-updating capability registry. The AI Assistant Capabilities note runs Dataview Serializer plugin for Obsidian queries that read SKILL.md frontmatter directly. No regeneration, no static snapshot. The receptionist's routing table cannot drift from the real fleet.
• Catalog visibility hygiene. Claude Code packs every visible skill's description into an 8000-char budget. Context loaders, *-shared and *-barrel skills MUST set disable-model-invocation: true. Routing precision survives growth instead of degrading with it.
• Progressive disclosure. Skills don't preload. Frontmatter is always visible; bodies only load when a skill actually fires. The receptionist routes from 250+ frontmatter blocks, then pulls one SKILL.md into context on demand. Pure Prompt Lazy Loading AI Design Pattern (PLL). Token budget stays flat as the fleet grows.
• Call-chain safety. Max depth 3, no duplicates, no cycles, chain context mandatory. Hard limits on emergent multi-agent behavior. Stops the "A calls B calls A" loop.

If you're building anything with more than ten AI skills, look at capability indirection and runtime path resolution first. Everything else is downstream of those two.

Get the Obsidian Starter Kit before the price goes up

Obsidian Starter Kit - Stop Configuring, Start Thinking | Knowledge Forge

Complete Obsidian vault with 40+ auto-filing templates, pre-configured plugins, and PKM methodology. 1,000+ users. 20+ years expertise. 30-day guarantee.

Knowledge ForgeDwayne Appleby

🔗 https://www.store.dsebastien.net/product/obsidian-starter-kit/

Related

• AI Assistant Architecture
• AI Assistant Capabilities
• AI Agent Skills
• Receptionist AI Design Pattern
• Prompt Lazy Loading AI Design Pattern (PLL)
• Claude Code
• Obsidian Starter Kit

About Sébastien

I'm Sébastien Dubois, and I'm on a mission to help knowledge workers escape information overload. After 20+ years in IT and seeing too many brilliant minds drowning in digital chaos, I've decided to help people build systems that actually work. Through the Knowii Community, my courses, products & services and my Website/Newsletter, I share practical and battle-tested systems.

I write about Knowledge Work, Personal Knowledge Management, Note-taking, Lifelong Learning, Personal Organization, Productivity, and more. I also craft lovely digital products and tools.

If you want to follow my work, then become a member and join our community.

Ready to get to the next level?

If you're tired of information overwhelm and ready to build a reliable knowledge system:

• 📚 KM for Beginners — 10+ hours of structured video lessons
• 🚀 Obsidian Starter Kit — Ready-made vault with 40+ templates
• 💼 Knowledge Worker Kit — Complete guides + lifetime community
• 🦉 1-on-1 Coaching — Personalized guidance
• 🎯 Join Knowii — Community + ALL courses & tools