routeledger logo

routeledger

Free

Which model actually served each turn of your Claude Code session, and where routing quietly diverged from what you configured. Read-only, offline.

Visit Website

At a glance

Explore the listed capabilities, then check the details that matter to you.

  • Audit Claude Code session transcripts
  • Compare declared and serving models
  • Detect model changes within sessions
ai-toolsagent-toolsanthropicauditclaudeaiai assistantautomation
routeledger screenshot 1

Overview

routeledger is a read-only command-line tool that audits Claude Code session transcripts to determine which model actually served each turn. It reads files already present under ~/.claude/projects and reports on model routing that the transcript records but does not surface. The tool runs three checks: declared model versus served model, model changes inside a session, and the plan-mode profile. Each check states its own limit.

Findings are reported in three states rather than two: ok, observation, and unverifiable. An observation is reported as a hypothesis with the reason it cannot be more than one, and unverifiable means the evidence needed is not in the file, so no verdict is issued. The tool records the Claude Code version behind every session it reads; on a version newer than the last one it was verified against, or when no version was recorded, findings drop to unverifiable.

routeledger makes no network calls, writes no files, cache, or telemetry, and does not change configuration. It reports no prompt or reply content, with one stated exception: the second check scans user text in memory for the /model command marker, and that value is passed to one regular expression and discarded. It requires Node 20 or newer.

Pricing

Free

Detailed plans have not been confirmed in our catalog. Check the official website for current limits and billing terms.

Visit Website

Prices and limits may change. Confirm the currency, billing period, seat minimum and usage caps on the official website.

Use Cases

  • Teams use routeledger to find out which model actually answered a Claude Code session.
  • Teams use routeledger to separate model changes they requested from changes they did not.
  • Teams use routeledger to compare the model a subagent run declared against the model that served it.
  • Teams use routeledger to check which models served plan-mode turns and which served the rest.
  • Teams use routeledger to obtain the same report as machine-readable JSON for further processing.
  • Teams use routeledger to list recent sessions so they can pick a session id.
  • Teams use routeledger to audit a specific session by id or id prefix.
  • Teams use routeledger with Node 20 or newer and transcripts under ~/.claude/projects, with no install, configuration, or account.

Features

Audit Claude Code session transcripts

Compare declared and serving models

Detect model changes within sessions

Inspect routing during plan mode

Mark findings with evidence limits

List and select recent sessions

Export reports as structured JSON

Reviews

No reviews yet. Be the first to share your experience!