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 dominikwozniak/dw-solo-skills --skill dw-handoffgit clone --depth 1 https://github.com/dominikwozniak/dw-solo-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/dominikwozniak/dw-solo-skills/dw-handoff)<a href="https://agentmods.dev/skills/dominikwozniak/dw-solo-skills/dw-handoff"><img src="https://agentmods.dev/badge/skills/dominikwozniak/dw-solo-skills/dw-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/dominikwozniak/dw-solo-skills/dw-handoff"><img src="https://agentmods.dev/badge/skills/dominikwozniak/dw-solo-skills/dw-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 | $0.00060 | $0.01025 |
| Opus 5 | $0.00030 | $0.00513 |
| Sonnet 5 | $0.00012 | $0.00205 |
| Haiku 4.5 | $0.00006 | $0.00103 |
Grade A, and why
dw-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 5d 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 — 102 lines — stays where its author put it; the contents beside it link to each section on GitHub.
dw-handoff — the middle of a task, saved
CHANGE.md already survives a /clear at task granularity. This skill covers the gap that
leaves — the middle of a task, where a checkbox cannot say which three approaches you already ruled
out — and holds only that: decisions belong in CHANGE.md, traps in ## Gotchas, follow-ups in
the backlog.
Output location
.ai/work/<date>-<slug>/HANDOFF.md, beside the CHANGE.md it describes. .ai/ is tracked in git.
Find the change by branch, the same way the rest of the loop does:
grep -l "^branch: $(git rev-parse --abbrev-ref HEAD)$" .ai/work/*/CHANGE.md 2>/dev/null
- One match — write beside it.
- Several — list them and ask.
- None — stop and write nothing. A handoff needs a change to attach to; say so and point at
dw-shape. An exploratory session with nothing shaped yet is what that skill is for. - Detached HEAD (the branch resolves to the literal
HEAD) — say so, list everyCHANGE.mdunder.ai/work/with its recordedbranch:, and ask which one this belongs to.
One live handoff per change. An existing HANDOFF.md is overwritten, never appended to — a
second handoff describes a newer moment, and keeping both leaves the next session guessing which one
is current. dw-next deletes it once it has ticked the task it described.
Workflow
1. Name the task in flight
Take the change from the grep above and, from its checklist, the task this handoff is about — usually the first unticked one, but confirm it against what the session actually did rather than assuming. State it in one line before writing anything, so a wrong guess is caught here.
2. Gather what the change doc can't hold
Four things, each grounded in something you can point at — a file:line, a command you ran, an error
you saw:
- How far into the task you actually are.
- What is applied but not committed — from
git status --shortandgit diff --stat, read, not recalled. - The dead ends already ruled out, each with the reason it failed. This is the line that earns the file: it is the work the next session would otherwise repeat.
- The single next concrete move.
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.
- 5d ago Changed · +2 lines e81de5b1f3e3
- 8d ago Changed 3e05b5de23ff
- 12d ago First seen · 100 lines · 60 tokens per session scan A 9cac1f112568
dw-handoff is a skill published in the GitHub repository dominikwozniak/dw-solo-skills (2 stars, last pushed today), licensed MIT. It adds 60 tokens to every session and 1,025 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 skills, from other repositories
distill
Retroactively distill the logs of finished work — delete what the repo now answers, keep what code cannot say. Done work only; in-flight logs are never touched.
memory-bank
Maintain and update project memory-bank files for persistent context across Claude Code sessions. Use when starting a new session, completing features, making architectural changes, or when context needs to be preserved. Triggers on: memory bank, project context, session start, context update, CLAUDE.md, project…
alive:session-history
Revive sessions (quick or heavy), browse, and search — 'what happened recently?', 'find the session where we discussed X', 'revive yesterday's session'. For single-session recall and multi-session browsing. If the human needs to merge multiple sessions into one working context or detect conflicts between parallel…
alive-mine
Nightly scan of session transcripts -- extract decisions, tasks, people, insights.
alive-people
Weekly -- cross-reference people mentions, nudge stale contacts.
alive-inbox
Scan 03Inbox/ for unrouted files, present routing suggestions.