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 ngreenwall/claude-skills --skill handoffgit clone --depth 1 https://github.com/ngreenwall/claude-skillsWrote 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/ngreenwall/claude-skills/handoff)<a href="https://agentmods.dev/skills/ngreenwall/claude-skills/handoff"><img src="https://agentmods.dev/badge/skills/ngreenwall/claude-skills/handoff/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/ngreenwall/claude-skills/handoff"><img src="https://agentmods.dev/badge/skills/ngreenwall/claude-skills/handoff.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 | $0.00111 | $0.03046 |
| Opus 5.5 | $0.00044 | $0.01218 |
| Sonnet 5.5 | $0.00022 | $0.00609 |
| Haiku 4.5 | $0.00011 | $0.00305 |
Grade A, and why
handoff 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 2mo 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 — 186 lines — stays where its author put it; the contents beside it link to each section on GitHub.
SKILL: Handoff
Leave durable project context for the next chat. Do not paste a full conversation recap or duplicate what already lives in project context files (CLAUDE.md, AGENTS.md, README.md, etc.).
When to run this
Not just end-of-task. Two triggers:
- Task done: run handoff, then start a new session anytime after, context window usage doesn't matter.
- Context running high, task not done: run handoff at a safe checkpoint (use the Next: field for what's left), then start a new session. Prefer this over
/compact, handoff is a reviewed, explicit capture;/compactis an automatic, lossy summary and risks silent drift. Reserve/compactfor a small amount of remaining work not worth a checkpoint.
PROTOCOL (run in this exact order)
0. Early exit for no-op sessions
Judge by session content first: did this session create or edit any files, or make any commits, including ones made mid-conversation outside this flow, and in prior handoff runs or a separate context-update pass earlier in the same session? If yes, there's work to log, continue from Step 1 regardless of what git status shows.
Only reach for git status --porcelain (terminal available, project is a git repo) as a confirmation check, and only once session content already points to nothing having happened, never as the primary signal. A clean working tree doesn't by itself mean nothing changed, earlier work may have already been committed and pushed outside the handoff flow, and that's real shipped work that still needs logging. Skip to Step 13 with a lightweight report only when session content says nothing happened: state that nothing changed this session, and mark every checklist line as no changes/already up to date.
If no terminal is available, or the project isn't a git repo, git status can't help at all, judge by session content alone.
1. Detect bootstrap gaps and route to project-init when needed
Check whether baseline scaffolding is missing:
CLAUDE.mdREADME.md.cursor/rules/context-router.md
What ships with it
3 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.
- 2mo ago First seen · 186 lines · 111 tokens per session scan A 07b7a8eac9b8
handoff is a skill published in the GitHub repository ngreenwall/claude-skills (2 stars, last pushed 2mo ago), licensed MIT. It adds 111 tokens to every session and 3,046 once invoked, about $0.0004 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
prismer-notion
Notion API + ntn CLI: pages, databases, markdown, Workers.
prismer-teams-meeting-pipeline
Teams meeting summaries, job replay, Graph subscriptions.
prismer-apple-reminders
Apple Reminders via remindctl: add, list, complete.
document-actions
Extract cited facts, obligations, deadlines and proposed actions from documents, preserving OCR uncertainty and reconciling Prismer tasks.
meeting-actions
Turn supplied meeting notes or transcripts into cited decisions, commitments, unresolved questions, and deduplicated Prismer tasks.
weekly-review
Review commitments and stalled work from Prismer tasks and memory, then produce an evidence-based weekly plan with explicit coverage gaps.