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 arimet/handoff --skill handoffgit clone --depth 1 https://github.com/arimet/handoffWrote 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/arimet/handoff/handoff)<a href="https://agentmods.dev/skills/arimet/handoff/handoff"><img src="https://agentmods.dev/badge/skills/arimet/handoff/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/arimet/handoff/handoff"><img src="https://agentmods.dev/badge/skills/arimet/handoff/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 | $0.00217 | $0.01721 |
| Opus 5.5 | $0.00087 | $0.00688 |
| Sonnet 5.5 | $0.00043 | $0.00344 |
| Haiku 4.5 | $0.00022 | $0.00172 |
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 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 — 155 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Handoff — verify you understood Claude's code
When Claude builds a feature, it is easy to merge the result without truly
reading or understanding it. handoff turns passive review into an active
check: it interviews the developer to confirm they read the code (where the
logic lives) and understood it (why decisions were made, how it behaves),
then reports a score and the spots to revisit.
This skill conducts an interactive quiz. Ask questions one at a time and wait for an answer before continuing. Conduct the entire quiz in the developer's language (match the language they are writing in).
Step 1 — Gather the material (hybrid source)
Build the quiz from two complementary sources:
-
Conversation context — what was built this session: the files touched, the key decisions and trade-offs, what was deliberately NOT done, and what could plausibly break. This is the source for the "why".
-
Git diff (when available) — to objectively anchor "where" and check that nothing important is missed. Run:
git rev-parse --git-dirto confirm a repo exists.git diff --statthengit diff(covers unstaged + staged) to enumerate changed files and lines. Use realfile:linelocations from this diff.
If there is no git repo, degrade gracefully and rely on conversation context alone — still produce a useful quiz.
Step 2 — Calibrate the number of questions
Scale to the size of the change (use the diff stat, or your best estimate from context if there is no git):
| Change size | Questions |
|---|---|
| Small (< ~50 lines, or 1–2 files) | 3 |
| Medium | 5 |
| Large (> ~300 lines, or ≥ 5 files) | 8–10 |
Step 3 — Generate a mixed question set
Cover both reading and understanding. Mix these types:
- Where (multiple choice). "Which file/function contains X?" Provide plausible distractors drawn from other real files in the change.
- Why (open-ended). "Why was approach X chosen over Y?", "What trade-off does this make?", "What would have broken if we hadn't done Z?"
- Behavior (multiple choice or open). "If input X arrives, what happens?", "How would you debug it if X fails?"
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.
- 1mo ago First seen · 155 lines · 217 tokens per session scan A 363e945f75f9
handoff is a skill published in the GitHub repository arimet/handoff (28 stars, last pushed 2mo ago), licensed MIT. It adds 217 tokens to every session and 1,721 once invoked, about $0.0009 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 2026-08-30.
Other skills, from other repositories
eli5
Explain any topic, code, concept, or error tailored to a specific audience's level of understanding. Use this skill whenever the user says 'explain like I am', 'ELI5', 'explain this to my', 'break this down for', 'dumb it down', 'simplify this for', or asks you to explain something to a specific person or audience…
wong
Compile everything Claude did today across ALL sessions and repos into one dated teaching-guide markdown saved in "/Daily Log", render it as a phone-readable PDF, and upload to Google Drive. Safe to run multiple times a day — it appends only newly-uncovered work to that day's single file rather than overwriting. Use…
youtube-course-builder
Use when the user wants to learn something, asks for a video, or says "teach me X", "I want to learn X", "how does X work", "explain X", "is there a video about X", "recommend a course/tutorial on X". Builds a progressive YouTube course through wide, creative search, reading the transcript and the comments before…
oral-english-practice
A Claude Code skill that stores and analyses spoken-English practice completed in the Claude app. The app handles the conversation; this skill keeps reports, scores, mistakes, progress trends, and the next practice focus.
game-programming-languages
Game programming languages - C#, C++, GDScript. Learn syntax, patterns, and engine-specific idioms for professional game development.
html-effectiveness-research
A guide for explaining how a feature works in a code repository or how an abstract technical idea works. It can structure the explanation with a summary, expandable steps, code views, questions and answers, or an interactive visual.