radin-stats

Files (2)

• SKILL.md
• openai.yaml · agents/

SKILL.md (first lines)

# Stats Roundup

Print each tool's own stats or gain command, back to back. Compute no merged
total. The units differ (per-session tokens, benchmark medians, a per-repo
ledger). Display each output as-is.

## Run every source

Run all four without probing. A source that is not installed fails visibly;
that failure is the skip, so do not retry it.

- **`/caveman:caveman-stats`** -- measured: real per-session token usage and
  savings, read from the session log itself.
- **`/ponytail:ponytail-gain`** -- benchmark: ponytail's published scoreboard.
  Not this session, and not this repo.
- **`/ponytail:ponytail-debt`** -- measured: ponytail's per-repo ledger of
  deferred shortcuts, if the repo has one.
- **`radin stats`** -- measured: skill, slash-command and sub-agent use on this device, per vendor, plus each radin repo's journal totals. Read from Claude Code transcripts, so it covers every companion. It also prints `rtk gain` and `caveman-stats.js --all`.

**Reply:** each source's output as-is, in the order listed, with no merged total.

From shortcuts/radin, MIT.

More from shortcuts/radin