Skip to content

jouet stats ​

  • Usage: jouet stats [FLAGS]

Show what jouet saved and cost: measured from its ledger, estimated from your history

Counts what the hook ledger recorded (jouet command runs, rule decisions and what the agent did after a denial, checks, and failing CI logs returned as excerpts), then estimates the input tokens those runs saved against your own Claude Code and Codex history from before jouet setup, and what jouet cost the same way: stop blocks, denials, failed jouet runs, jouet edit failures past the old rate, and the primer on every request since setup. The net may be negative. Every estimate shows its basis, and a kind of work with fewer than 20 runs before setup is not estimated. Without a ledger, it shows what one run of each command would save.

Flags ​

  • --since <TIME> — Only ledger entries since TIME: a duration ago such as 7d or 12h, a date such as 2026-10-01, or RFC 3339
  • --harness <HARNESS> — Which harness: claude, codex or all (default all)
  • --session <ID> — Only one session's ledger entries
  • --summary — Print one line about what jouet saved, or what it cost when that was more, or nothing, for a stop hook. [stats] summary in the config file or JOUET_STATS_SUMMARY sets its style: quips (default), plain or off
  • --rescan — Read the history before setup again instead of reusing the baselines the last run saved
  • -h --help — Print help
MIT LicenseCopyright © 2026jdx.dev