# Context Budget

> Keep the agent's context lean so accuracy doesn't collapse. Use on long sessions, big files, or when the agent starts hallucinating.

- Skill: `elitongadotti/context-budget` (Agent Skill)
- Install (CLI): `npx skillmds@latest add elitongadotti/context-budget`
- Raw SKILL.md: https://api.skillmd.com/api/skills/elitongadotti/context-budget/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: elitongadotti (https://skillmd.com/u/elitongadotti)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/elitongadotti/context-budget

---

# Context Budget
More context is not better. Past a threshold, accuracy falls off a cliff (context rot).
- **Don't dump whole files** — read the relevant function/section, not the 2000-line module.
- **Summarize, don't accumulate** — replace a 200K-token transcript with a 4K recap of the load-bearing facts.
- **Drop dead context** — once a sub-task is done, its detail leaves the window. Keep the decision, drop the trace.
- **State on disk, not in context** — progress goes to a file the next turn re-reads, not into an ever-growing prompt.
A bloated standing context (CLAUDE.md) taxes every single turn. Trim weekly. If the agent is confidently wrong, suspect the context before the model.