Walsworth-Group/physkit / skills/handoff
handoff
VerifiedCold-resumable handoff for a fresh agent: goal, state and in-flight reasoning, a confirm-state command, next steps. Use before pausing or a context reset.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 8.2 KB |
| assets/handoff_template.md | markdown | 2.0 KB |
assets/handoff_template.md
View on GitHub# Handoff — {{PROJECT}} — {{UTC_DATE}}
> The live handoff. A fresh agent with only this file and the repo must understand the work, trust its
> state, and take the next step. Prior handoffs live in `archive/`; never overwrite them.
## 1. Vision & goal
- **End goal:** <what the work must achieve, concretely>
- **The decision it drives / why it matters:** <one paragraph>
- Full goal: `campaign.md` §Mission · what to try next: `results/JOURNAL.md`
## 2. Current state (incl. in-flight reasoning)
- **Current best / result:** <value + where it lives (config/snapshot)> · **verification:** <its label: in a campaign the checks artifact's (e.g. self-verified, CONFIRMED/peer); else CONFIRMED / PROVISIONAL / unverified> · **metric epoch:** <metric_version · standing; any pending epoch awaiting ratification>
- **Done:** <the settled pieces>
- **In-flight, and why:** <the hypothesis being tested, why, what was expected, what had just been learned>
- **Open threads:** <live questions>
- **Landmines / dead ends:** <known-broken things; directions already ruled out and why (→ `campaign.md` §Known landmines / §Decision record)>
## 3. Run this to confirm state
```bash
# environment
<activate env, e.g. source .venv/bin/activate>
# correctness gates (expect: all pass / suite GREEN)
python scripts/metric.py --selftest
python scripts/run_campaign.py --preflight # names the metric epoch + its standing
<run the correctness-suite, e.g. make check>
# reproduce the current best (expect: <STATED VALUE> ± tol)
<command that re-derives the current-best number from its snapshot>
```
**Expected output:** <what a healthy state prints, so a mismatch is obvious>
## 4. Roadmap / NEXT (prioritized, goal-linked)
1. <next step> · *serves the goal by …* · <enough detail to act on>
2. <…>
3. <…>
Kept in sync with the JOURNAL's **NEXT:** line and §What to try next and why.
---
<!-- On the next handoff: copy this to archive/handoff_<this date>.md, then write a fresh dated file. -->
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |