Skills over MCP plus progressive disclosure: fixing context bloat in domain-specific AI systems
sebgehr · x · 2026-10-22
Domain-specific AI systems need repeatable processes that LLMs can't be trusted to get right, but hard-coding everything bloats context. Seb Gehr's engineering blog argues for Skills over MCP plus progressive disclosure: load instructions and tools on demand to keep both reliability and context lean.
More from coding & agent
- swyx hails Opus 5.5 as best work-estimator model; dev counters that review time is the real bottleneck — petrusenko_max · 2026-10-23
- Six defense kits released for AI agents after Railway and LiteLLM supply-chain incidents — alvelda · 2026-10-23
- Agent Field Kit launches free agentic defense toolkit against LLM-enhanced attackers — alvelda · 2026-10-23
- LiteLLM PyPI compromise and Railway breach: Oct 21 attack timeline and free response kit — alvelda · 2026-10-23
- Daily Experiment Planner agent: one prompt to save scientists hours of lab prep — nlarusstone · 2026-10-23
- Two £300 Snapdragon mini PCs, no GPU: 46 tok/s local Qwen3.8 Flash Next with custom ToRc engine — WorldlyAd7946 · 2026-10-22