Getting it into your agent
This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.
/plugin marketplace add cwakim/claude-plugins/plugin install memory-backupWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/commands/cwakim/claude-plugins/share)<a href="https://agentmods.dev/commands/cwakim/claude-plugins/share"><img src="https://agentmods.dev/badge/commands/cwakim/claude-plugins/share/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/commands/cwakim/claude-plugins/share"><img src="https://agentmods.dev/badge/commands/cwakim/claude-plugins/share.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00078 | $0.03101 |
| Opus 5.5 | $0.00031 | $0.01240 |
| Sonnet 5.5 | $0.00016 | $0.00620 |
| Haiku 4.5 | $0.00008 | $0.00310 |
Grade A, and why
share scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 238 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Share (share <thread>)
Packages one thread of work so a colleague, or you on a machine with no
access to the backup repo, can pick it up. Unlike zip, which mirrors
everything for your own disaster recovery, share is curated and
sanitized: it carries only what someone else needs to continue a single
thread, with your personal memories held back by default and every file
secret-scanned before it leaves the machine. It does not require /backup setup to have run.
The bundle uses the same machines/<hostname>/ layout as every other
command (see ${CLAUDE_PLUGIN_ROOT}/docs/layout.md), so the recipient
ingests it with the restore or merge they already have, no new command
on their side. The scan behavior is in
${CLAUDE_PLUGIN_ROOT}/docs/secret-scan.md. Read both before building a
bundle.
share is interactive-only, by design. Every step is a judgment call:
which thread, which repos, which plans, whether a held-back memory should
go, where the bundle lands. There is no cron or headless mode. If share
is ever invoked with nobody to ask (for example from a non-interactive
script), log the problem and exit cleanly rather than guessing, the same
as zip and merge.
Step 1 - Resolve the thread and its scope
<thread> is required.
- Match it against
~/.claude/handoff-index.md, case-insensitively and allowing partial matches, against the titles and one-line goals in the index (the same matching/pickupuses). This file is written by the session-continuity plugin;sharereads it as a plain file and there is no hard dependency.- One live match (its path still exists): that handoff note is the anchor.
- Several: list them and ask which via
AskUserQuestion. - No match, but
<thread>looks like a path (contains/or ends.md, expand a leading~): use that file as the anchor if it exists. - No match and no index: say so and stop. Suggest running
/handoffon the thread first, sinceshareis built around a handoff note.
- Derive the in-scope set from the anchor note:
- The anchor handoff note itself, plus a one-line index entry for it (see Step 3).
- The repo the note lives in: from the note's home-relative path
(
~/sites/foo/.claude/handoff.md-> repo~/sites/foo), derive its memory store slug perdocs/layout.md(sites-foo->~/.claude/projects/$(echo ~ | tr '/' '-')-sites-foo/memory/). If that store exists and is non-empty, it is in scope. - Other repos named in the note body: scan the note text for other
absolute or
~-relative repo paths. For each whose memory store exists, ask viaAskUserQuestionwhether to include that store too. Never pull in a store the note only mentions in passing without asking. - Plans: if
~/.claude/plans/exists and is non-empty, list every plan file and ask viaAskUserQuestion(multi-select) which belong to this thread. Handoff notes do not reliably link plan files, so this is never automatic; default the selection to none.
- If the derived scope is empty (no memory store, no plans, just the handoff note), say so and confirm the user still wants a handoff-note-only bundle before continuing.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 238 lines · 78 tokens per session scan A 5a8369343ae2
share is a command published in the GitHub repository cwakim/claude-plugins (2 stars, last pushed 23d ago), licensed MIT. It adds 78 tokens to every session and 3,101 once invoked, about $0.0003 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-11.
Other commands, from other repositories
ralph
Ralph autonomous development loop with fresh context per iteration.
create-prd
Create PRD and set up project for Ralph autonomous development loop.
optimize
Run the credo optimisation audit for the current repo: a freshness check, then a read-only scan (conflict hotspots, changelog fragments, test-stage convention, dogma settings, parallelism readiness, and - only if the user says Yes when asked before the scan - instruction consistency across the global and repo…
gws-tasks
List, create, complete, and manage Google Tasks.
gws-standup
Generate a standup report combining today's calendar and tasks.
milepost
Force a milepost milestone entry for the current session now.