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 skills add ApurvBazari/claude-plugins --skill savegit clone --depth 1 https://github.com/ApurvBazari/claude-pluginsWrote 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/skills/apurvbazari/claude-plugins/save)<a href="https://agentmods.dev/skills/apurvbazari/claude-plugins/save"><img src="https://agentmods.dev/badge/skills/apurvbazari/claude-plugins/save/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/skills/apurvbazari/claude-plugins/save"><img src="https://agentmods.dev/badge/skills/apurvbazari/claude-plugins/save.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 | $0.00062 | $0.01600 |
| Opus 5.5 | $0.00025 | $0.00640 |
| Sonnet 5.5 | $0.00012 | $0.00320 |
| Haiku 4.5 | $0.00006 | $0.00160 |
Grade A, and why
save 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 — 126 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Save Skill — Capture Session Intent for Handoff
You are auto-invoked when the user signals end-of-session, or invoked explicitly via /handoff:save. Capture what the next session needs to pick up, then write it to .claude/handoff/active.md.
Step 1: Determine invocation mode
| Trigger | Mode |
|---|---|
User typed /handoff:save (explicit) |
Direct mode — skip the confirm step in Step 2; proceed straight to Step 3 |
| Auto-invocation on NL trigger phrase | Confirm mode — run Step 2 before anything else |
If unsure, default to Confirm mode — a single AskUserQuestion is always safer than a silent false-positive save.
Step 2: Confirm before saving (Confirm mode only)
Ask the user via AskUserQuestion (single-select, 3 options):
- Save now (Recommended) — "I'll draft a directive from the current session and write it to
.claude/handoff/active.mdimmediately." - Edit before saving — "I'll draft a directive, show it to you, let you edit, then write."
- Cancel — "Don't save. I'll continue the current task."
If Cancel: exit silently. Do not write anything. If Save now or Edit before saving: proceed to Step 3.
Step 3: Synthesize the directive
Draft a directive (3-6 bullets or short prose) capturing what the next session must know to pick up cleanly. Pull from the current conversation context.
Include:
- What to pick up — the concrete next step (one sentence, action-oriented).
- Pointers — memory files, key source paths, git refs, PR numbers worth reading first.
- Constraints — what NOT to do (decisions already made, scope guards, "don't relitigate X").
- State expectations — what should be true at resume (e.g., "PR #50 is open against develop, do not push").
- Open questions — if any are unresolved and the next session needs to handle them.
Style rules:
- Directive prose only. Do NOT embed state that the next session can derive — git log will tell them current commits / branch / working-tree status; memory files will tell them prior context. Capture intent, not snapshots.
- Use present-tense imperative addressing the future Claude instance: "Read
<memory>first", "Do NOT push without confirming". - Length budget: under 400 words. If you need more, the directive is doing too much.
- If the user is in the middle of a multi-step task with a written plan, point to the plan rather than restating it.
What ships with it
2 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 126 lines · 62 tokens per session scan A 827363464a38
save is a skill published in the GitHub repository ApurvBazari/claude-plugins (3 stars, last pushed 8d ago), licensed MIT. It adds 62 tokens to every session and 1,600 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
c-notify
Send native macOS notification center alerts from the command line using terminal-notifier.
anti-sleep
Keep a Mac awake with caffeinate during long builds, downloads, or supervised automation runs.
onlyswitch-deeplink
Invoke OnlySwitch built-in switches or buttons via deeplink. Use when the user asks to toggle or trigger an OnlySwitch action (e.g. empty trash, keep awake, dark mode, mute, hide desktop, clear clipboard, eject discs, Xcode derived data, and any other built-in switch or button). Resolve the request to the correct…
gmail
Manage Gmail email — drafting, sending, organizing, filters, vacation replies, and inbox analysis.
outlook
Manage Outlook email — drafting, sending, organizing, rules, vacation replies, and inbox analysis.
watcher
Create and manage polling watchers that monitor external services (Gmail, Google Calendar, GitHub, Linear, Outlook) for events and process them with custom action prompts.