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 goldenwo/claude-state-drift --skill handoffgit clone --depth 1 https://github.com/goldenwo/claude-state-driftWrote 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/goldenwo/claude-state-drift/handoff)<a href="https://agentmods.dev/skills/goldenwo/claude-state-drift/handoff"><img src="https://agentmods.dev/badge/skills/goldenwo/claude-state-drift/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/goldenwo/claude-state-drift/handoff"><img src="https://agentmods.dev/badge/skills/goldenwo/claude-state-drift/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 | $0.00073 | $0.00566 |
| Opus 5 | $0.00036 | $0.00283 |
| Sonnet 5 | $0.00015 | $0.00113 |
| Haiku 4.5 | $0.00007 | $0.00057 |
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 10d 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.
What it actually says
handoff
End the session cleanly: state.json catches up, then a narrative handoff is composed and parked for the next session. The NEXT session's orientation auto-loads it (same project only) until a newer handoff or state write supersedes it.
Procedure
- Update state first. Run the full
update-stateskill contract (draft → show diff → WAIT for approval → write →state-validate→state-historyappends). If nothing material changed, say so and skip the write. - Compose the narrative. Write the handoff body in THIS session's voice — the things state.json deliberately excludes. Use exactly this shape:
# Handoff — <date>: <one-line theme>> **TL;DR:**2-3 sentences: where things stand + the single honest next move## Where we are— commits/branches/PRs landed this session, working-tree state## What's NEXT— the next move and why; an honest STOP if nothing validated remains## What's LEFT— blockers, flakes, dangling threads, don't-re-attempt items## Resume notes— gotchas, how-to-verify, carry-forward context Keep it under ~5KB — the orient render truncates at 6KB.
- Write it. Save the body to a temp file, then:
state-handoff write --body-file <tmpfile>(session id defaults fromCLAUDE_CODE_SESSION_ID). Confirm the tool's output line to the user. - Tell the user the handoff is parked and will auto-load in the next session for this project.
Anti-patterns
- Composing from the transcript summary instead of your live understanding (compaction-grade output).
- Writing the handoff WITHOUT the state update — the supersession rule keys on state.json freshness.
- Padding with content already in state.json (deliverable lists, version history) — the orient block carries those.
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.
- 10d ago First seen · 31 lines · 73 tokens per session scan A ef188bda359c
handoff is a skill published in the GitHub repository goldenwo/claude-state-drift (1 stars, last pushed 23d ago), licensed MIT. It adds 73 tokens to every session and 566 once invoked, about $0.0004 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
recall
Must be used at the start of any non-trivial task involving code changes, debugging, repo exploration, file inspection, or environment/tooling investigation to surface stored guidance before analysis or tool use.
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…
cross-session-handoff
Read, write, snapshot, and lock .arcgentic/state.yaml across planner, dev, audit, and optional test sessions.
knowledge-priming-refiner
Facilitate a structured conversation to create a project-specific knowledge base document. Produces a knowledge-base.md that primes AI with the project's tech stack, architecture, trusted sources, and project structure. Use when the user says 'set up knowledge base', 'prime the project', 'onboard AI', 'create…
alive-inbox
Scan 03Inbox/ for unrouted files, present routing suggestions.
alive-mine
Nightly scan of session transcripts -- extract decisions, tasks, people, insights.