Agent instruction bloat

Instructions for coding agents, such as AGENTS.md guidance, process skills and workflows, tend to grow longer over time. Although each new rule might be useful in isolation, this accumulation often leads to agent instruction bloat. The instructions become long, contradictory and stale, especially if teams engage in agent skill cognitive surrender and import external instruction sets wholesale.

First, convert any instructions you can into deterministic hooks or checks. Then, reevaluate instructions against each new model and delete those it no longer needs; progression testing, rather than regression testing. Anthropic removed over 80% of Claude Code's system prompt for its latest models with no measurable loss, as frontier models now follow instructions more reliably. Progressive context disclosure keeps the remaining instructions out of the way until they’re needed.

Context files such as AGENTS.md and CLAUDE.md tend to accumulate over time as teams add codebase overviews, architectural explanations, conventions and rules. While each addition is useful in isolation, this often leads to agent instruction bloat. Instructions become long and sometimes conflict with each other. Models tend to attend less to content buried in the middle of long contexts, so guidance deep in a long conversation history can be missed. As instructions grow, the likelihood increases that important rules are ignored. We also see many teams using AI to generate AGENTS.md files, but research suggests that hand-written versions are often more effective than LLM-generated ones. When using agentic tools, be deliberate and selective with instructions, adding them as needed and continuously refine toward a minimal, coherent set. Consider leveraging progressive context disclosure to surface only the instructions and capabilities an agent needs for its current task.