agentleFS
Sign inStart free
eltonylfgi-blipeltonylfgi-blip/claude-code-usage-guard/skills/usage/SKILL.md

usage

Skill4 starsChanged 4 months agoMITView on GitHub

Show your Claude Code usage right now — your real plan quota (5h + weekly) when available, plus a weighted-spend proxy, burn rate, and how close you are to your budget. Use when the user asks "how much have I used", "usage status", "am I burning fast", "cuánto llevo gastado", "/uso".

6 other projects use a version of this.

Adopted by

Most recently taken up by

  • Other versions6 projects use a version of this.

    Word for word · 0

    None found.

    Edited · 0

    None found.

    Wrote their own version · 6

    1. jdx/usage★ 1.1ksame name
    2. crisandrews/ClawCode★ 64same name
    3. sachingulati/claude-statusline★ 2same name
    4. Miyamura80/MCP-Template★ 1same name
    5. Kakhramon/usage-analyzer★ 0same name
    6. wlodarczyk-pawel/claude-usage-monitor★ 0same name

    How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.

  • In this repositoryOne other file it gives its agents.

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 3 are distinctive, and 1 is very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 3

    Distinctive

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If the status-line shim has been wired up (see README §3.5), the output LEADS with a

    line 19
  2. Distinctive:

    If the "Real plan quota" block is present, report those percentages and reset countdowns as the headline.

    line 25
  3. Very specific:

    If a config file exists at ~/.claude/usage-guard.json with a weightBudget, also compare the weighted spend against it (e.g.

    line 29
  4. Distinctive:

    If usage is high or the burn rate is climbing, give ONE concrete tip to slow the spend (e.g.

    line 38

“Common” is measured: kept by at least 6 in 10 of the 4 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Usage status

Report the user's current Claude Code usage by reading the locally-captured plan quota and their local session transcripts.

Steps

  1. Run the engine:

    node "${CLAUDE_PLUGIN_ROOT}/lib/engine.mjs"
    • If the status-line shim has been wired up (see README §3.5), the output LEADS with a "Real plan quota" block: the actual 5-hour and weekly used_percentage and reset times. This is the number that truly gates the user — report it first.
    • Below that it prints the weighted-spend proxy for the last 5 hours: burn rate, output/input/cache tokens, turn count, and models used.
  2. If the "Real plan quota" block is present, report those percentages and reset countdowns as the headline. If it is ABSENT, tell the user real quota isn't being captured yet and that they can enable it by wiring the status line (README §3.5) — until then, fall back to the proxy.
  3. If a config file exists at ~/.claude/usage-guard.json with a weightBudget, also compare the weighted spend against it (e.g. "72% of your 8M fallback budget"). If no budget is set and no real quota is captured, say so and point to the two ways to enable warnings.
  4. Present a short, plain-language summary (define any jargon in-line). Example shape:
    • Plan quota: 5h 88% used (resets in 1h 29m) · weekly 41% — when available
    • Spent (last 5h, proxy): 4.5M weighted · burn rate ~1.2M/h
    • Breakdown: output 183k · fresh input 417k · cache 3.1M created / 8.4M read
  5. If usage is high or the burn rate is climbing, give ONE concrete tip to slow the spend (e.g. start a fresh session to drop the cache-read tail, batch questions, avoid re-reading large files). Keep it to one line — do not lecture.

Notes

  • Real plan quota comes from Claude Code's rate_limits (5h + weekly), captured by the status-line shim to ~/.claude/.usage-guard-limits.json. It's the actual plan limit, not a guess. Present only for Claude.ai Pro/Max after the first API response.
  • "Weighted spend" is the FALLBACK proxy: input + output + cache-creation at face value, cache-read counted at 0.1 (it is ~10x cheaper). A proxy, not an exact bill and not the plan quota — used when real rate_limits aren't available.
  • The engine has zero dependencies and only reads local files; it never makes network calls.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free