weigh-bags
Record scale readings for each bag — empty (tare), outbound, current, final — and reconcile them against the itemised inventory to find unaccounted weight. Use when the user gives a bag weight, says how heavy something is, weighs their luggage, or asks how close to the limit they are. Appends to bags.yaml and runs scripts/reconcile.py.
- License 许可
- MIT
- 最近更新
- 2026-11-10
- Stars
- ★ 0
- Forks
- 0
01 前言 Front Matter
- name
- weigh-bags
- description
- Record scale readings for each bag — empty (tare), outbound, current, final — and reconcile them against the itemised inventory to find unaccounted weight. Use when the user gives a bag weight, says how heavy something is, weighs their luggage, or asks how close to the limit they are. Appends to bags.yaml and runs scripts/reconcile.py.
- allowed-tools
- Read, Write, Edit, Bash(uv run *), Bash(ls *), Grep
03 SKILL.md 正文
Weigh Bags
$TPA below is this plugin's new-workspace skill directory — the folder holding its SKILL.md,
scripts/, schema/, typst/ and template/. Find it by name in your tool's skills folder (some
tools prefix it, e.g. travel-packing-assistant__new-workspace) and set TPA once per session.
Turns scale readings into the numbers the plan is checked against, and finds the gap between what the inventory says a bag contains and what it actually weighs.
The four phases
| phase | what it is | why it matters |
|---|---|---|
empty |
the bag on the scale with nothing in it — its tare | every allowance is gross weight, so tare comes straight off the usable capacity |
outbound |
what it weighed leaving home | the baseline the return leg is measured against |
current |
where it stands now, mid-trip | the working number for a re-pack |
final |
the check-in reading | the ground truth; record it, it is what makes the next trip's estimates good |
Readings append. Never overwrite an earlier weighing — the history across a multi-leg trip
is the data, and a bag that gained 4 kg between outbound and current is telling you
something a single current reading cannot.
Getting a reading the user can trust
- Hook/luggage scale — right tool. Lift smoothly and read at rest; a jerked lift reads high by a kilogram or more. Take two readings and use the second.
- Bathroom scale, by difference — weigh the person, then the person holding the bag, subtract. Works, but bathroom scales typically quantise to 0.1–0.5 kg and the error applies twice, so treat the result as ±1 kg and leave the headroom to match.
- Kitchen scale — for items, not bags.
Whatever the method, ask what the resolution is once and record it in scales[]. A reading of
"23" from a scale that rounds to the nearest kilogram is not the same fact as 23.0 kg from a
scale that reads to 100 g, and the difference is exactly the size of the margin people get
caught by.
Reconciliation
After recording, run:
uv run $TPA/scripts/reconcile.py . --json
It computes, per bag, gross − tare − cube tare − itemised contents = unaccounted, and
classifies:
- consistent — the inventory explains the bag.
- partial inventory — a real but modest gap. Normal. Say what it is in kilograms and treat the measured weight as authoritative for planning, not the itemised sum.
- inventory incomplete — the gap is large enough that the plan cannot be trusted. The fix is to add the missing items, not to add a fudge factor.
When the residual has the same sign and similar size on every bag, that is the signature of
a miscalibrated scale rather than missing items; reconcile.py will suggest a bias_kg. Show
the suggestion, explain what it means, and let the user decide whether to write it into
bags.yaml. Never apply a calibration correction silently — a systematic offset applied to a
correct scale produces a plan that is wrong on every bag at once.
Which number the plan uses
Where a bag has both a measured gross weight and an itemised total, the measured weight wins, and the difference is carried as unaccounted mass. That is the conservative choice: it is the number the airport scale will agree with.
Procedure
- Read
bags.yaml. If a bag has noemptyweighing, ask for its tare before anything else — without it every downstream figure is off by the weight of the suitcase. - Record the readings the user gives, with the phase, the date and the scale id.
- Run
reconcile.py. - Report per bag: gross, allowance, headroom, status, and the unaccounted mass. Lead with the arithmetic — gross minus allowance — before any interpretation.
- If any bag is over, or inside 1.5 kg of its limit, name it and offer
/travel-packing-assistant:planto redistribute or/travel-packing-assistant:overageto price the alternatives.
04 同仓库更多 Skills
no-description
暂无简介
good-skill
Fixture skill that satisfies every rule.
broken-link
Links to a reference file that does not exist.
Bad_Name
Name violates the kebab-case rule and does not match the directory.
skill-authoring
Write, review, and validate portable agent skills in SKILL.md format. Use when creating a new skill, editing an existing one, or asked whether a skill follows the Agent Skills spec.
packing-inventory
Build or update the inventory of what is being carried — items, quantities, weights, carriage restrictions, and how replaceable each thing is at the destination. Use when the user is listing what they have packed or intend to pack, dictating the contents of a bag, or asking what is in their luggage. Writes inventory.yaml, which the solver and the leave-behind ranking both read.