Limitlens — AI skill for Claude Code
Local-first quota intelligence for Codex and Claude: cache-adjusted token attribution, per-model breakdown, and reset-aware pacing.
How to install Limitlens
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open ahmtsahin/limitlens and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Limitlens does
Local-first quota intelligence for Codex and Claude: cache-adjusted token attribution, per-model breakdown, and reset-aware pacing.
Alternatives in AI
- Codeburn — Free, local tool to track AI coding token usage and cost across 37 tools and agents (Claude Code, Cursor, Code 9.7k ★
- Token Monitor — Local-first desktop widget for tracking token usage, costs, and limits across 32+ AI coding tools—including Cl 1.8k ★
- TokenTracker — Local-first AI token usage & cost tracker for 31 coding tools incl 1.4k ★
README
LimitLens for Codex & Claude
**Project-level quota attribution and reset-aware pacing — 100% local.**
Know which repository used your Codex or Claude quota, how much coding capacity is left, and whether your current pace reaches the limit before the window resets.
[](https://www.npmjs.com/package/limitlens) [](https://www.npmjs.com/package/limitlens) [](LICENSE) [](package.json) [](https://github.com/ahmtsahin/limitlens/actions/workflows/ci.yml) [](#known-limitations)
npx --yes limitlens@latest
Requires **Node.js ≥ 22.13**, Git in `PATH`, and local Codex and/or Claude Code session metadata. Open the printed loopback URL, normally .
The walkthrough uses clearly labelled synthetic documentation data. The command reads supported metadata on your own machine; LimitLens does not upload it.
Why LimitLens
- Ranks observed work, not cache noise. Cache-adjusted token totals keep re-read context from dominating project rankings.
- Identifies repositories reliably. Normalized Git remotes merge duplicate checkouts while keeping unrelated same-name folders separate; per-turn model attribution survives model switches.
- Forecasts honestly against reset windows. Provider-reported quota truth stays separate from reconstructed local history, and sparse data is labelled
Related Skills
Auto Model Router
Cache-, quota- and capability-aware LLM router with a Claude Code shim
Contextdiet
Analyze what eats your AI agent's context window — token breakdown, trim suggestions, local-first
Token Station
Local routing control plane for AI agents and LLM providers, with a loopback-only gateway, smart and quota-awa
Cab
One local gateway for every coding agent. Cost-aware LLM model routing for Claude Code, Codex, OpenCode, and m
Vibe Gauge
🧠 Native macOS menu-bar dashboard for AI coding: Claude / Codex / Gemini / Grok quotas with sleep-aware usage
VibeGauge
🧠 Native macOS menu-bar dashboard for AI coding: Claude / Codex / Gemini / Grok quotas with sleep-aware usage
Related Agents
Stats Analyzer
Analyzes ~/.claude/stats-cache.json and history.jsonl to produce a concise usage report with model token break
Kimi Worker
Forward a coding, diagnosis, or review task to Kimi K3 (Moonshot) through the local kimi-worker.sh runtime. Us
Feature Estimator Breakdown
Sub-agent of the feature-estimator pipeline. Produces the task breakdown table with AI-adjusted estimates and