Claude Code usage
Golden Thread
Cuts Claude Code token cost: handoffs to fresh sessions (20–66% lower cost, 8 test runs) and LOTR, which keeps your MCPs from eating your tokens (measured on 9 bulk reads: median 74% less context growth, 45–94% on 8 of 9). Plus shared markdown memory and hook-enforced rules. MIT.
Install
/plugin marketplace add shaven/golden-threadFound in the project README. Check it for setup steps after installing.
git clone https://github.com/shaven/golden-thread.gitFrom the README
Golden Thread Reader: someone who has never heard of Golden Thread Stamped for gt 0.22.1: the counts and version numbers in this document are checked against the code at every release; the prose is re-read when the document is edited, not at every release. A memory system for AI coding sessions, built on plain markdown and git — and, unusually, one where the rules that matter most are mechanically enforced rather than merely written down. Every AI session starts with amnesia. You explain your conventions, the session does good work, the session ends, and it is all gone. Golden Thread is one Obsidian vault that acts as shared memory between you and every session you run: sessions read from it at startup, look things up while working, and write back what they learn. Its distinguishing idea is the second problem, the one most memory systems never address: writing a rule down does not mean it gets followed. Plugin v0.22.1. Ten Core rules currently enforced, five of them *validated* — a hook inspects the finished reply (Stop) or the tool call about to run (PreToolUse) and blocks it if the rule was broken. The commands The skills are how you talk to Golden Thread in a session. These are the tools it installs into your vault, at Projects/golden-thread/tools/, which you or a skill run directly. They exist because some things must not depend on a model remembering to do them. log.md, decisions.md and TASKS.md are generated. Hand-editing them is not a style violation — your change is simply lost at the next merge, and gt-lint reports it. The checks, and when each one runs These live with the plugin rather than in your vault, at ~/.claude/plugins/cache/golden-thread-plugin/gt//scripts/, because a git hook and a launchd job have to find them without a vault being open.…
Details
- Repository
- shaven/golden-thread
- Language
- Python
- License
- MIT
- Last updated
- 5 November 2026
FAQ
What is Golden Thread?
Cuts Claude Code token cost: handoffs to fresh sessions (20–66% lower cost, 8 test runs) and LOTR, which keeps your MCPs from eating your tokens (measured on 9 bulk reads: median 74% less context growth, 45–94% on 8 of 9). Plus shared markdown memory and hook-enforced rules. MIT. It is a usage for Claude Code, from shaven/golden-thread on GitHub.
How do I install Golden Thread?
Run /plugin marketplace add shaven/golden-thread inside Claude Code, then follow the setup in the README.
Is Golden Thread popular?
shaven/golden-thread has 27 GitHub stars and 3 forks. It was last updated on 5 November 2026.
More usage & cost trackers
Claumon
17fabioconcina/claumon
Claude Code dashboard: see where your limits are headed before you hit them. Calibrated usage forecasts, live rate-limit gauges, session costs, memory browser. Single binary, zero config, plus a VS Code extension.
Checkyouragent
26TarekAwwad/checkyouragent
An analytics tool to explore Claude code usage patterns, errors, and token costs
Aireceipts
31anandgupta42/aireceipts
Your AI coding agent just billed you. Here's the receipt. Cost receipts for Claude Code, Codex, Cursor, Gemini & opencode — no account, no upload.
Retok
33d-date/retok
Token-efficiency analyzer for AI coding agents (Claude Code, OpenAI Codex CLI): cost estimates, cache analysis, and savings advice. Zero dependencies.
Battery
27allthingsclaude/battery
A macOS menu bar app that monitors your Claude Code API usage in real time.
Model Citizen
25JakeSelby/model-citizen
The control plane for your coding agents, however you run them. One checkout of rules, skills, subagents, hooks and stances, projected into Claude Code and Codex with your own primitives: hook-enforced guardrails, usage telemetry, cost postures and model tiering, under any rules library or orchestration layer.
Browse the most popular Claude Code mods, all usage & cost trackers, or the skills and MCP server directories.