agentleFS
alexjmiller5alexjmiller5/agent-config-public/skills/quota/SKILL.md

quota

Skill0 starsChanged 43 days agoMITView on GitHub

Use when checking how much of an AI coding subscription is left - Claude Code, Codex, Cursor, Copilot, Grok, Kimi, Z.AI or Antigravity plan quota, rate-limit windows, reset times or burn pace - when the user asks "how much do I have left", or before deciding whether it is safe to keep spending a provider's quota.

2 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

1 rule. About 0 are common, 0 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    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. Very specific:

    Never run quota-axi update on a package-manager install (read-only store) - bump the package version instead.

    line 52

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

quota (quota-axi)

Binary: quota-axi (kunchenguid/quota-axi), installed via the machine's package manager (nix here). Data only: it reads local provider credentials and first-party usage endpoints; it never routes, ranks, or mints tokens. quota-axi --help is the syntax authority over this file.

Commands

NeedRun
Everything, compact TOON (default for agents)quota-axi
One or a few providersquota-axi --provider claude,codex
Normalized model for jqquota-axi --provider claude --json
Per-window pace / reserve / account evidencequota-axi --full
Which credential sources exist (no secret values)quota-axi auth
Human live dashboard (needs a real tty - not from an agent shell)quota-axi --tui

Reading the default report

  • quota[] - one row per measurable scope: effectivePercentRemaining, resetsAt (ISO, UTC), runway / limitedBy (which window binds first), spendPriority.
  • exhaustion[] - only scopes projected to run out before reset, with the ETA.
  • attention[] - anything non-nominal: auth_required, error, unresolved_windows, plus a remedy column. Read this block before reporting numbers - an empty quota[] with an attention row means "not measured", not "0% left".
  • Exit 0 even when some providers failed; 1 = every provider failed; 2 = usage error.

Gotchas

  • macOS Claude quota needs a one-time Keychain grant. Until then Claude shows auth_required · keychain_access_required. The USER runs quota-axi --allow-keychain-prompt once and clicks Always Allow - never click that dialog for them. The ACL attaches to /usr/bin/security, and the grant marker is per Claude account, so it survives package rebuilds.
  • Codex with cli_auth_credentials_store = "keyring" keeps its login in the OS keychain, so quota-axi auth shows auth-json missing - that is fine as long as cli-rpc is available: quota is read through codex app-server. A Codex quota unavailable error with a working codex login status means the probe itself broke (Codex >= 0.149 rejects the -a untrusted flag upstream still passes; run codex -s read-only -a never app-server by hand to confirm, patch the flag at install time until upstream #177 lands).
  • Any other provider that is installed but not logged in reports error / missing sources - the fix is the vendor's own login, not quota-axi.
  • A plain read may delegate an expired token's refresh to the vendor CLI; pass --no-credential-refresh when the read must stay strictly read-only.
  • Never run quota-axi update on a package-manager install (read-only store) - bump the package version instead.
  • Percentages are per provider; never present one provider's % as comparable to another's.

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.