Benjamin Thillerup
Community@bthillerup · Copenhagen
Agent Skills by Benjamin Thillerup
Showing 13 vetted skills indexed across 1 GitHub repositories.
bthillerupmulti-agent-patterns
Design and implement multi-agent systems with architectural patterns and context isolation.
bthillerupbdi-mental-states
Models agent beliefs, desires, and intentions using formal BDI ontology patterns.
bthillerupclean-code
Enforce pragmatic coding standards for concise, direct code.
bthillerupcontext-degradation
Diagnose context-length performance degradation in large language models.
bthillerupcontext-compression
Compress large conversation histories and codebases exceeding context window limits.
bthillerupmemory-systems
Design and implement persistent agent memory systems across sessions.
bthillerupadvanced-evaluation
Implement LLM-as-a-Judge direct scoring and pairwise comparison for model output evaluation.
bthillerupcontext-fundamentals
Educates users on AI context engineering principles and optimization strategies.
bthillerupproject-development
Guide LLM project development from task-model fit to pipeline architecture.
bthillerupevaluation
Evaluate AI agent performance using rubrics and degradation testing.
bthillerupcontext-optimization
Compress conversational history and mask tool outputs to reduce token usage.
bthillerupfilesystem-context
Offload large tool outputs to files for dynamic context discovery.
bthilleruptool-design
Design and refine agent tools with clear descriptions and consolidated architecture.