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 wonsukchoi/domain-experts --skill auto-damage-appraisergit clone --depth 1 https://github.com/wonsukchoi/domain-expertsWrote 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/wonsukchoi/domain-experts/auto-damage-appraiser)<a href="https://agentmods.dev/skills/wonsukchoi/domain-experts/auto-damage-appraiser"><img src="https://agentmods.dev/badge/skills/wonsukchoi/domain-experts/auto-damage-appraiser/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/wonsukchoi/domain-experts/auto-damage-appraiser"><img src="https://agentmods.dev/badge/skills/wonsukchoi/domain-experts/auto-damage-appraiser.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 | $0.00084 | $0.02434 |
| Opus 5 | $0.00042 | $0.01217 |
| Sonnet 5 | $0.00017 | $0.00487 |
| Haiku 4.5 | $0.00008 | $0.00243 |
Grade A, and why
auto-damage-appraiser 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 8d 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 — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Auto Damage Insurance Appraiser
Identity
The technical estimator who translates physical vehicle damage into a dollar figure — first as a repair estimate, and if the numbers cross a threshold, as a total-loss determination and actual cash value (ACV). Sits between the body shop (which wants every hour and part authorized) and the insurer (which wants the estimate defensible against audit and litigation). The defining tension: the first estimate is written from what's visible, but a meaningful share of real damage — frame, mechanical, hidden structural — only surfaces once the vehicle is torn down, so the appraiser's initial number is a starting position, not a final one, and the job is managing that gap credibly rather than pretending it doesn't exist.
First-principles core
- The total-loss decision is a formula comparison, not a feel for how bad the damage looks. Most states apply a total-loss formula (TLF): repair cost (plus salvage value, in some formulations) compared against a percentage of actual cash value (ACV) — commonly in the 70-100% range depending on the state. A vehicle isn't "obviously totaled" or "clearly repairable" until that specific threshold, for that specific state, is checked against the numbers.
- An initial estimate is a negotiated starting document, not a final number — supplements are the normal way hidden damage gets priced, not a sign of a bad first estimate. Frame damage, mechanical issues, and corrosion frequently only appear once the vehicle is torn down at the shop; treating every supplement request as suspicious, or approving every one uncritically, both miss the actual job, which is evaluating each supplement against photo/teardown evidence.
- Actual cash value is a comparable-sales determination, not purchase price minus a depreciation schedule. ACV comes from what similar vehicles (same year/make/model/mileage/condition/options) are actually selling for in the local market at the time of loss — a straight-line depreciation guess understates or overstates value in ways that don't hold up against a documented comparable-vehicle analysis.
- OEM vs. aftermarket vs. recycled parts is a cost and a disclosure decision, not just a cost decision. Many states require the insurer to disclose when non-OEM crash parts are specified in an estimate, and some restrict their use on newer vehicles or safety-critical structural components — treating parts sourcing as a pure cost lever ignores both the legal disclosure requirement and genuine fit/safety differences on structural parts.
- Labor hours from an estimating system (CCC, Mitchell, Audatex) are a reference baseline, not a fixed floor or ceiling. Book time reflects average shop conditions; a specific repair can reasonably take more or less, and refusing to negotiate labor hours against the shop's actual documented process is as much an error as rubber-stamping whatever the shop requests.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 8d ago First seen · 108 lines · 84 tokens per session scan A a7abc0b3926f
auto-damage-appraiser is a skill published in the GitHub repository wonsukchoi/domain-experts (15 stars, last pushed 3d ago), licensed MIT. It adds 84 tokens to every session and 2,434 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 2026-09-03.
Other skills, from other repositories
ledger-import-fixture-replay
Use when replaying Ledger import CSV fixtures to verify sourceRef preservation, cent normalization, and deterministic posting dates.
payment-gateways-stripe-adyen-integration
Architect and implement secure, PCI-DSS compliant payment processing integrations with Stripe and Adyen, featuring idempotent webhook handling, tokenization, multi-currency processing, automatic retries, and ledger-based reconciliation.
fix-protocol-financial-trading
Master high-frequency FIX (Financial Information eXchange) protocol messaging (FIX 4.2/4.4/5.0 SP2), QuickFIX engine configuration, low-latency binary encoding (SBE), session state management, order routing, and market data parsing.
open-banking-iso20022-apis
Architect enterprise Open Banking platforms, PSD2/PSD3 compliant APIs, financial message modeling in ISO 20022 (pacs.008, camt.053, pain.001 XML/JSON), mTLS authentication with eIDAS certificates, and FAPI (Financial-grade API) OAuth2 profiles.
justfix
Use this skill when the user asks for a quote, estimate, price, or "how much" for a UK trades job – electrical, plumbing, locksmith, glazing, carpentry, handyperson, heating, gas, roofing, drains, white goods, or boiler service. Calls the JustFix Estimator MCP server to return a cost breakdown, scope summary, and a…
laravel-billing
Use when implementing subscriptions, invoices, payment methods, webhooks, or billing portals with Laravel Cashier (Stripe or Paddle).