context-budget
Overview
Audit what's eating my Claude Code context — skills, MCP servers, the CLAUDE.md chain, and
my growing memory — and produce a prioritized list of what to trim. The point is keeping
the whole setup lean, not just the pack. This is skill #11, a conscious break of the
original 10-cap (since raised again, 11→12, for exploit-hunt), justified by token economy
being a recurring concern.
When to use
- Context feels heavy / output quality slips, or
/context shows a big always-on chunk.
- I've added skills, agents, or MCP servers, or
MEMORY.md keeps growing.
- Before adding more — to know if there's headroom.
The key model: always-on vs on-demand
- Always loaded (every turn): skill
description lines, MCP tool schemas, the CLAUDE.md
chain, and MEMORY.md (the memory index). This is the real budget — trim here first.
- On-demand (only when invoked): a skill's SKILL.md body, an agent's prompt. Cheap — a
long body costs nothing until it runs, so don't optimize bodies.
How it works
1. Inventory
- Skills:
make details (or claude plugin details <plugin>) for the pack's projected
token cost; note each description length (always-on).
- MCP servers: count servers and tools (~500 tokens of schema per tool, always-on). Flag
servers with many tools, or that just wrap a CLI I already have (
gh, git, supabase).
- CLAUDE.md + memory: size of the CLAUDE.md chain and
MEMORY.md + memory/*.md. The
memory index is always-on, so its growth is a direct context tax.
- Agents: description length (loads on every spawn) and body size.
2. Classify each component
- Always needed → keep.
- Sometimes needed (domain-specific, rarely triggered) → lazy-load / disable until used.
- Rarely needed / redundant / overlapping → remove.
3. Detect the usual bloat
- Skill/agent
description over ~30 words (an always-on tax on every turn).
- MCP over-subscription (tools I never call, or CLI-wrapper servers).
MEMORY.md entries that are stale, duplicated, or no longer true → prune them.- CLAUDE.md verbosity that should be a rule or a memory instead.
4. Report
Total always-on overhead, a breakdown by component (skills / MCP / CLAUDE.md / memory /
agents), the headroom left, and a prioritized trim list — biggest always-on saving first.
Output
A short report plus the top 3 trims by always-on tokens saved. Act on the always-on items
first; on-demand bodies are not the problem.
1---2name: context-budget3description: Use when context feels heavy or you've added skills, agents, MCP servers, or memory and want a token-consumption audit of the whole setup with a prioritized trim list.4---56# context-budget78## Overview910Audit what's eating my Claude Code context — skills, MCP servers, the CLAUDE.md chain, and11my growing memory — and produce a prioritized list of what to trim. The point is keeping12the **whole setup** lean, not just the pack. This is skill #11, a conscious break of the13original 10-cap (since raised again, 11→12, for `exploit-hunt`), justified by token economy14being a recurring concern.1516## When to use1718- Context feels heavy / output quality slips, or `/context` shows a big always-on chunk.19- I've added skills, agents, or MCP servers, or `MEMORY.md` keeps growing.20- Before adding more — to know if there's headroom.2122## The key model: always-on vs on-demand2324- **Always loaded** (every turn): skill `description` lines, MCP tool schemas, the CLAUDE.md25 chain, and **`MEMORY.md`** (the memory index). This is the real budget — trim here first.26- **On-demand** (only when invoked): a skill's SKILL.md body, an agent's prompt. Cheap — a27 long body costs nothing until it runs, so don't optimize bodies.2829## How it works3031### 1. Inventory3233- **Skills:** `make details` (or `claude plugin details <plugin>`) for the pack's projected34 token cost; note each `description` length (always-on).35- **MCP servers:** count servers and tools (~500 tokens of schema per tool, always-on). Flag36 servers with many tools, or that just wrap a CLI I already have (`gh`, `git`, `supabase`).37- **CLAUDE.md + memory:** size of the CLAUDE.md chain and `MEMORY.md` + `memory/*.md`. The38 memory index is always-on, so its growth is a direct context tax.39- **Agents:** description length (loads on every spawn) and body size.4041### 2. Classify each component4243- **Always needed** → keep.44- **Sometimes needed** (domain-specific, rarely triggered) → lazy-load / disable until used.45- **Rarely needed / redundant / overlapping** → remove.4647### 3. Detect the usual bloat4849- Skill/agent `description` over ~30 words (an always-on tax on every turn).50- MCP over-subscription (tools I never call, or CLI-wrapper servers).51- `MEMORY.md` entries that are stale, duplicated, or no longer true → prune them.52- CLAUDE.md verbosity that should be a rule or a memory instead.5354### 4. Report5556Total always-on overhead, a breakdown by component (skills / MCP / CLAUDE.md / memory /57agents), the headroom left, and a **prioritized trim list — biggest always-on saving first**.5859## Output6061A short report plus the top 3 trims by always-on tokens saved. Act on the always-on items62first; on-demand bodies are not the problem.
Endika