
Context Budget Auditor
Audits how much of Claude Code's context window your agents, skills, MCP servers and rules consume, and suggests savings.
- Repository
- affaan-m/ECC
- Stars
- 295,127
- Licence
- MIT
- Files
- 1 file
What it does
This skill makes the agent act like a auditor for your Claude Code setup. Each session loads instructions, tool descriptions and project notes that take up room in the model's limited working memory (the context window). When invoked, the agent inventories those components, estimates their token cost using rough word-count formulas, flags bloated or duplicated items, and produces a ranked report with suggested cuts such as removing unused MCP servers or lazy-loading rules.
Great for
- Developers who have installed many skills, agents or MCP servers and notice sessions feel slower or less accurate
- Power users planning to add more MCP servers and wanting to check there is still room
- Teams maintaining shared CLAUDE.md files and rule sets who want to trim redundancy
Skip it if
- People using Claude apps or other tools without Claude Code's agents/skills/MCP folder structure, since the scans target those paths and may find nothing useful
- Anyone expecting exact token counts: the skill estimates with words x 1.3 or characters / 4, so figures are approximate
Try asking
- “/context-budget”
- “Audit my Claude Code setup and tell me which MCP servers I should remove to free up context.”
- “I want to add 5 more MCP servers, do I have room in my context window?”
- “Run a verbose context budget report and show me the heaviest agent files.”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the MIT licence and an attribution note.
context-budget.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip context-budget.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip context-budget.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Run the audit after adding agents, skills or MCP servers to catch context creep early.
- Use the verbose mode only when debugging a specific bloated file, since the full output is long.
- Treat the reported token numbers as estimates and verify large savings before deleting components.
What's inside
- SKILL.md
Redistributed unmodified from affaan-m/ECC under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.