Claude Code Plugin

/plugin marketplace add empathic/toolpath
/plugin install path@toolpath

Ask your session history anything

Every session you run is recorded somewhere on disk, in a format you can't read, by a tool that never looks at it again. /path:query lets you question that history in plain English — and not just Claude Code's: it reads Codex, Gemini CLI, GitHub Copilot, opencode, Cursor, and Pi sessions too, so you can ask about one harness or across all of them.

/path:query what did i work on last week?

Most of last week went to the payments service: migrating off the legacy webhook handler (Tue–Thu, 640 steps) and two short sessions chasing a race in the retry queue. Monday was the docs-site redesign; Thursday afternoon a one-off backfill script. No recorded activity Friday or over the weekend.

/path:query which sessions cost me the most tokens this month?

The heaviest was Wednesday's session on the import-pipeline bug: roughly 8.4M tokens — 7.9M of it cache reads, 61k generated output. Worth knowing: the top two sessions are segments of the same conversation; together they're nearly half the month's total.

/path:query compare my claude and codex usage this month

Claude Code carried most of the volume: 9.2M output tokens across 31 sessions, against Codex's 480k across 7. Codex runs shorter sessions — a median of 41 steps to Claude's 220 — and spends about 4% of its output on reasoning. Claude doesn't itemize thinking tokens, so its reasoning share isn't measurable.

/path:query what parts of my workflow were inefficient this month?

Two things dominate. About a fifth of the month's steps were later rewound or replaced, concentrated in three sessions that each had to be substantially redone. And marathon sessions are where the tokens go: the five sessions over a thousand steps account for nearly half your context spend — a fresh session per sub-task would be the highest-leverage change. Tool failures barely register by comparison (2.3% of calls, mostly shell friction).

/path:share

This conversation, as a link. Publishes the current session to Pathbase. Name another session in plain words ("the one about the flaky test") and it finds and shares that one instead.

/path:share the session where we fixed the retry queue

Found it — Tuesday's session on the retry-queue race. Shared: pathbase.dev/u/you/pathstash/retry-queue-race

/path:link-pr

Ship the why with the diff. Shares the session and appends the link to your pull request's description, so the review carries the conversation behind the change. Saying "share this conversation to the PR" is enough to trigger it:

share this conversation to the pr

Shared — the session is at pathbase.dev/u/you/pathstash/rate-limit-retry and PR #212's description now links to it.

/path:resume

Sessions are portable. Point it at a Pathbase URL and it projects the session into your project, ready to pick up with /resume — even if the session started on another machine, or in a different harness entirely.

/path:resume pathbase.dev/u/mira/pathstash/retry-queue-race

Imported into this project. Run /resume 7c9e2b41 to pick up where Mira left off.

How it works

On first use the plugin installs the path CLI from a GitHub release (sha256-verified) — or uses the one already on your PATH. The CLI reads each harness's on-disk session format and normalizes it into Toolpath documents in a local cache; /path:query answers questions from that cache entirely on your machine. Sharing is explicit: only /path:share and /path:link-pr upload anything, and you choose anonymous or signed-in uploads.