Instead of trying to make agentic delegation deterministic by setting keyword variables as a "YOU MUST INVOKE X-AGENT IF THE USER SAYS DELEGATE", keywords = a score, a certain score = a specific system prompt that is injected into the session post UserPromptSubmit. You can read how this works more on the repo: (https://github.com/blas0/klaus-baudelaire)
Delegates built-in agents + some redefined ones (broad general roles + hyper-specific tool callers) Some agents were built from/around Claude's exposed system prompts and whatever invisible native agents they're hiding behind the scenes.
Bonus: optional memory manager tuned to how Claude actually thinks (not some vector DB cope) to help with managing memory based on: https://code.claude.com/docs/en/memory - doing away with the branded ".memory" directories and such.
Zero external APIs, zero services, zero reinventing (besides context7).
Feels like what Anthropic probably wanted us to do before everyone started LARPing as OpenAI. Plugin version drops late Monday–early Wednesday (this week, Jan 2026).
Video coming day-of with routing/delegation demo + bare Claude Code vs Klaus Baudelaire.
Thoughts? Go ahead and roast or don't! Please, just don't make another 500-line prompt wrapper pls and call it a novel abstraction.
I have been building NTNT (/in'tent/), an experimental programming language exploring a specific question I have had:
"What would a programming language look like if we assumed AI agents would be the primary writers of code instead of humans?"
The core concept is Intent-Driven Development (IDD). Instead expressing intent through vibe coding or markdown file chains, you (and your agent) define a machine-verifiable .intent file (using IAL - Intent Assertion Language) that serves as an executable contract between human and agent. The agent then maintains, builds and verifies with .intent file driving development.
The language itself is "Rust-flavored scripting" designed to reduce hallucinations and ease strain on agents as they author and manage code.
I'd love your feedback:
1. Does the NTNT flavor of IDD have potential to improve agent based development flows? 2. If I/we extracted the Intent system to work with TypeScript/Go, would you use it?
Repo: https://github.com/ntntlang/ntnt Website: https://ntnt-lang.org (runs on ntnt)