Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/cwakim/claude-plugins/zipgit clone --depth 1 https://github.com/cwakim/claude-pluginsWhat 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 | $0.00054 | $0.01081 |
| Opus 5 | $0.00027 | $0.00541 |
| Sonnet 5 | $0.00011 | $0.00216 |
| Haiku 4.5 | $0.00005 | $0.00108 |
Grade A, and why
zip 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 yesterday.
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 — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Zip (zip <path>)
A portable, dated archive of everything a backup would mirror, without
GitHub: no repo, no gh, no network. Useful for a USB copy, an air-gapped
machine, or a one-off snapshot before wiping this laptop. Does not require
/backup setup to have run.
The mirror tree and naming rules live in
${CLAUDE_PLUGIN_ROOT}/docs/layout.md; the scan behavior in
${CLAUDE_PLUGIN_ROOT}/docs/secret-scan.md. Read both before building the
archive.
Steps
<path>is required and must be an existing directory (expand a leading~); if it is missing or not a directory, say so and stop. There is no default destination: writing an archive is a deliberate act.- Ask via
AskUserQuestion, every run, before touching anything: scan the mirrored files for secrets before writing the archive, or skip the scan?- Scan (recommended): same secret-scan behavior as a GitHub backup.
- Skip: copy everything as-is, no scanning, no redaction. Warn
plainly, once, before proceeding: the archive may then contain
credentials verbatim, and it is only as safe as
<path>itself (a bare USB stick or an unencrypted folder is not a private GitHub repo). This choice is per-run and is never remembered; a zip made for quick, local carrying is a different threat model than the one it makes for someone who's about to hand<path>to another machine or drive.
- Build the same tree a backup run would, in a fresh temp directory
(
mktemp -d), undermachines/<hostname>/for parity with the GitHub layout and the manual restore fallback: the full inclusion list, naming rules, and exclusions are indocs/layout.md, and the handoff-copying rules (index parsed with python, archived handoffs collected separately) are identical to a backup run's. - If scanning was chosen, scan every file for secrets per
docs/secret-scan.md. Reuse this machine's shared.redact-allowread-only if it exists; new findings are asked interactively (batched, same three choices) and includes go to a.redact-allowbundled inside the archive itself, never the shared one, per the allowlist section ofdocs/secret-scan.md. If scanning was skipped, do nothing here: every file goes in as found. - Write
manifest.jsonat the archive root perdocs/layout.md, with"scanned": true/falserecording which choice step 2 made. - Zip the temp directory into
<path>/memory-backup-<hostname>-<YYYY-MM-DD-HHMM>.zip, then remove the temp directory. - Report: the archive's full path and size, counts mirrored per store, and whether the scan ran. If scanning ran: every redaction or omission (loudly, same as a backup). If it was skipped: repeat the warning from step 2 once more here, so it isn't just a prompt someone clicked past.
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.
- yesterday First seen · 77 lines · 54 tokens per session scan A 45adce6acfe8
zip is a command published in the GitHub repository cwakim/claude-plugins (2 stars, last pushed 11d ago), licensed MIT. It adds 54 tokens to every session and 1,081 once invoked, about $0.0003 per session on Opus 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 2026-08-31.
Other commands, from other repositories
checklist
Generate a custom checklist for the current feature based on user requirements.
clarify
Identify underspecified areas in the current feature spec by asking up to 5 highly targeted clarification questions and encoding answers back into the spec.
specify
Create or update the feature specification from a natural language feature description.
analyze
Perform a non-destructive cross-artifact consistency and quality analysis across spec.md, plan.md, and tasks.md after task generation.
converge
Assess the current codebase against the feature's spec, plan, and tasks, then append any remaining unbuilt work as new tasks to tasks.md so implement can complete it.
implement
Execute the implementation plan by processing and executing all tasks defined in tasks.md.