feat(skill): lab cost structure + pricing — full cost stack before quoting a price

[areliw]

What

Closes a coverage gap in two draft judgment skills: labs routinely price on direct cost alone (reagents/consumables) and underprice, ignoring indirect/hidden cost that can equal direct cost.

skills/lab-clinic-business-judgment.md (primary home)

• Fork 7 — build the full cost stack before quoting a per-test price / judging viability. The direct-vs-hidden landmine (de-identified restaurant analogy), the 4 hidden-cost buckets:
  1. Fully-loaded headcount (salary + benefits + leave/absence coverage), not base salary
  2. Rent + utilities + internet
  3. Service/repair contracts + non-lab staff (cleaner, aircon tech on rounds)
  4. Equipment depreciation on capital purchases
  • Decision rule: true unit cost ≈ variable/test + (fixed/month ÷ tests/month) → price ≥ true unit cost even for non-profit / government.
  • The government-ignores-fixed-cost dynamic ("the state pays for it") that suppresses reference prices and is one structural reason MT wages stay low.
• Fork 8 — price-floor dynamics. Why a cartel/price-fix breaks (the low-cost 5th player always undercuts; also illegal), and why government procurement is the biggest price-suppressor.
• Fork 6 unit-economics bullet now points to Fork 7; 3 new anti-patterns.

skills/finance-judgment.md

• Fork 8 — general true-unit-cost bullet for pricing your own service + one anti-pattern, cross-linked to the lab skill (principle only, no duplication — its Fork 2 reads others' financials, so the gap was distinct).

Why

Community landmine from a Thai MT group expert (de-identified), analogous to a well-known restaurant that miscalculated cost and failed. Frames as decision-support, not a cost-accounting lecture — the judgment is "before you quote a price, build the full fixed+variable stack and check price ≥ true unit cost."

Reviewer notes

• Both skills remain status: draft (ADVISE, not clinical 🩸) — no A/B gate required by CI.
• Numbers/structures hedged as teaching illustrations that vary by unit — verify against real accounts. PII-clean.
• Regenerated skills/INDEX.md + dist/all-skills.md via build_triage.py; validate_repo.py passes (96 skills, frontmatter + links OK). prompts/triage.md unchanged (taglines didn't move).

🤖 Generated with Claude Code

[chatgpt-codex-connector]

💡 Codex Review

Here are some automated review suggestions for this pull request.

Reviewed commit: d1b52c49d1

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[chatgpt-codex-connector]

Qualify full-cost price floor for subsidized work

For labs with spare capacity or fixed costs already committed/subsidized, work priced below fully allocated average cost can still improve contribution margin when the price exceeds variable cost; volume only makes losses worse when price is below variable cost or triggers new fixed costs. The unconditional ราคาขาย ≥ true unit cost เสมอ rule would make the skill reject viable government/non-profit or incremental contracts, so please frame this as the long-run/full-cost sustainability floor and add the variable-cost contribution exception.

Useful? React with 👍 / 👎.