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 aiya000/dotfiles --skill read-handoffgit clone --depth 1 https://github.com/aiya000/dotfilesWrote 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/aiya000/dotfiles/read-handoff)<a href="https://agentmods.dev/skills/aiya000/dotfiles/read-handoff"><img src="https://agentmods.dev/badge/skills/aiya000/dotfiles/read-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/aiya000/dotfiles/read-handoff"><img src="https://agentmods.dev/badge/skills/aiya000/dotfiles/read-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 | $0.00072 | $0.01129 |
| Opus 5.5 | $0.00029 | $0.00452 |
| Sonnet 5.5 | $0.00014 | $0.00226 |
| Haiku 4.5 | $0.00007 | $0.00113 |
Grade A, and why
read-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 — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
read-handoff
Picks up where the last session stopped. The user should not have to supply a path.
The last session left two files, not one, both in the ~/.ai-memory/ git repository: the handoff
under handoff/, and a memory file that the handoff names. Both get read, in that order. The
repository is shared by the local machine and Claude Code cloud sessions, so the handoff may have
been written on the other side.
First: the issue overview
Before looking for the handoff, run the show-issues-overview skill and show its table.
It is the one thing here that reflects the world as it is now. The handoff and the memory were both written in the past, so the overview is what they get checked against -- an Issue the handoff calls 未解決 may already be closed, and something new may have been opened since.
- Run it with no arguments -- the open issues are what matters at the start of a session
- If it cannot produce a list (no forge, not authenticated, no TODO file), do not stop -- say so in one line and carry on to the handoff. The overview is context, not a gate
Finding the file
-
Get the repository level with the
sync-memoryskill (prepare) -- a pull, or in a cloud session a clone. Without it, a handoff written on the other machine is not there yet. The directory is~/.ai-memory/handoff/. Handoffs written before the move to the repository live in the old local places,~/tmp/claude-handoff/and${TMPDIR:-/tmp}/claude-handoff/: look there too, and take the newest across all of them. Ifpreparesays~/.ai-memoryis not set up, use the old places alone and say so -
Work out the project the same way it does:
common=$(git rev-parse --path-format=absolute --git-common-dir 2>/dev/null) project=$( [ -n "$common" ] && basename "$(dirname "$common")" || basename "$PWD" ) -
Take the last of
<project>-*.mdsorted by name -- the names carry the timestamp -
If there is none for this project, take the newest
*.mdin the directory and say which project it was written for before acting on it -
If there is nothing at all, say so and list what the directory does hold. Do not guess
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 Changed e4b36ca56f0e
- 1mo ago Changed · +8 lines 988279d12b9a
- 1mo ago Changed · +13 lines 5d043ce3b929
- 1mo ago Changed · +15 lines · +8 tokens per session d0e21a098191
- 1mo ago First seen · 45 lines · 64 tokens per session scan A 6f75a4a5c8ca
read-handoff is a skill published in the GitHub repository aiya000/dotfiles (21 stars, last pushed 16d ago), licensed MIT. It adds 72 tokens to every session and 1,129 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 2026-09-07.
Other skills, from other repositories
use-things
Read and update the user's Things 3 to-dos, projects and areas with the things CLI (things-cli). Use whenever the user mentions Things, an item, project or area in Things, or asks to record findings or progress there. Read before every things command, reads included.
wrap-up-session
Answer "is this session all done?" by accounting for every action the session raised and listing what is still open. Run it in the message that starts a review session, and whenever the user asks whether the session can be closed.
use-rex
Rex gotchas, and the one way to start a Claude Code session in a Rex tab. Read before every rex command, because a tab started the wrong way stalls at a permission prompt nobody sees, and a failed step in a chained command can leave you reading your own pane. Re-invoke each time.
rex
Use the rex CLI to manage Rex sessions, open tabs, arrange panes, run commands, inspect terminal output, and control connected Rex apps.
use-notion
Reference for using the Notion API via the notion-agent subagent. Use to expore and update Notion workspace content.
restart__pull_request
A pull-request restart workflow for replacing a messy review with a new pull request containing one clean commit and a summary of the earlier discussion.