Skip to content

Add TokenWise to Monitoring & Analytics#62

Open
memurcie wants to merge 1 commit into
LangGPT:mainfrom
memurcie:add-tokenwise
Open

Add TokenWise to Monitoring & Analytics#62
memurcie wants to merge 1 commit into
LangGPT:mainfrom
memurcie:add-tokenwise

Conversation

@memurcie
Copy link
Copy Markdown

Summary

Adds TokenWise under Monitoring & Analytics in both `README.md` and `README-CN.md`.

What it does

A Claude Code skill that auto-routes subtasks across Haiku, Sonnet, and Opus based on task class, then logs every routed task to a local NDJSON with real token + cost numbers. Includes an A/B test subcommand that runs the same task across multiple tiers and scores quality.

  • Routes: Haiku for grunt work, Sonnet for scoped reasoning, Opus only for synthesis
  • Measures: every routed task logged with real $ numbers, zero telemetry, local-only
  • Proves: `/tokenwise:ab` validates the cheaper tier before you trust it
  • Privacy: no analytics endpoint, file contents never logged

Why this section

Fits naturally alongside `Claude-Code-Usage-Monitor`, `claude-code-costs`, `ccusage`, and `claude-code-otel` — all cost/usage tools. TokenWise is the only entry that routes while measuring (others only observe).

Files updated

  • `README.md` — added row under `## Monitoring & Analytics`
  • `README-CN.md` — added row under `## 监控与分析`

License: MIT

TokenWise — measurement-driven model router for Claude Code. Routes
Haiku/Sonnet for grunt work, reserves Opus for synthesis, logs every
routed task with real $ numbers. Added rows to both README.md and
README-CN.md under Monitoring & Analytics / 监控与分析.

Repo: https://github.com/CodeShuX/tokenwise
License: MIT
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant