Claude Usage Limits
claudeReduce Claude Code token usage without reducing output quality. Compact-code decoding, a patch tool that names a region instead of quoting it, delegated Glob/Grep search, MCP relocation and a CLAUDE.md splitter.
Six skills that make a Claude Code session fit inside your usage limits: put the 5-hour and weekly windows in your status line, count what loads on every turn, delegate verbose work to subagents, slim down CLAUDE.md, sharpen vague requests, and write repetitive files as compact notation. Free and MIT; the code is in this repository.
Plugin claude-usage-limits-optimizer from marketplace culp in emerickchasse/claude-code-plugin-marketplace-culp. Commands read from the repo's manifest — 1 month ago.
This plugin runs code on your machine — its hooks fire automatically on events like tool use. The exact commands are listed below; read them before installing.
count-tokensMeasure how many tokens a file costs Claude, and what share of the context window it occupies. Use this whenever the user asks how big their CLAUDE.md is, whether a file is too long for context, what is eating their context window, how many tokens something costs, or before advising anyone to shorte
delegate-to-subagentsDecide whether to delegate work to a subagent, and pick the right agent type and model when you do — so verbose output stays out of the main context instead of being paid for on every later turn. Use this whenever a task will produce output you will not reference again (test runs, log files, wide se
optimize-claude-mdRestructure a CLAUDE.md so the tokens it costs on every single turn are the ones that earn it — move reference material into on-demand skills, keep the standing rules. Use this when a CLAUDE.md has grown past a few thousand tokens, when the user asks how to write or shorten a CLAUDE.md, when a fresh
optimize-promptTurn a vague request into a specific brief before doing the work, so a wide scan and a discarded first attempt never happen. Use when a request names no file, symbol or error — "improve this codebase", "add tests", "make it faster", "clean this up", "review my code" — or when the user asks to optimi
show-usage-limitsPut the user's Claude usage limits in their status line — the 5-hour and weekly windows, how much is spent, and when each resets — so hitting a limit stops being a surprise. Use this whenever someone asks to see their usage, wants to monitor rate limits or context usage, says they keep hitting limit
stretch-usage-limitsDiagnose why a Claude Code session is burning through usage limits and fix the causes, in order of what actually costs the most. Use this whenever the user says they hit their limit, ran out of usage, got rate limited, wants to know why Claude is "so expensive", asks how to make their plan last long
write-compact-codeWrite a new Python or JavaScript file as compact Emmet-like notation that a bundled script expands into real source, so the repetitive structure never has to be emitted token by token. Use this when creating a file with repetitive shape — many similar methods, a batch of route handlers, parallel dat
prompt-optimizerTurn a vague request into a specific, actionable brief in CO-STAR form, naming the concrete targets it should have named and listing what is genuinely unknown as questions rather than inventing answers. Use before starting broad work such as "improve this codebase", "add tests", "make it faster" or
python3
Reduce Claude Code token usage without reducing output quality. Compact-code decoding, a patch tool that names a region instead of quoting it, delegated Glob/Grep search, MCP relocation and a CLAUDE.md splitter.
Talk like caveman. Cut filler. Keep technical details explicit.
22 skills · 1 command · 3 agents · hooks
Persistent memory for Claude Code: the next session starts where the last one left off.
29 skills · 1 MCP server · hooks
tasteskill 2.5.1: professional company websites at the figma.com and x.ai level, with one section per screen, real photographs, real product UI, effects kept for tech and gaming pages, and a real-browser checker and finish gate.
1 skill
Open-source AI job search that runs locally in your AI coding CLI: scan job portals, evaluate listings into a structured A-H report with a 1-5 score, tailor your CV, track applications
5 skills · hooks
Cross-session memory and token savings for coding agents.
6 skills · 1 agent · 1 MCP server · hooks
Package any website or local web build into a lightweight desktop app with Pake CLI.
1 skill
This page updates itself from your repo. Add the badge to your README, or get it Featured to pin it above the organic results with a followed link to your site.