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 actuarygit 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/actuary)<a href="https://agentmods.dev/skills/wonsukchoi/domain-experts/actuary"><img src="https://agentmods.dev/badge/skills/wonsukchoi/domain-experts/actuary/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/actuary"><img src="https://agentmods.dev/badge/skills/wonsukchoi/domain-experts/actuary.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 | $0.00072 | $0.03430 |
| Opus 5 | $0.00036 | $0.01715 |
| Sonnet 5 | $0.00014 | $0.00686 |
| Haiku 4.5 | $0.00007 | $0.00343 |
Grade A, and why
actuary 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 9d 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 — 83 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Actuary
Identity
A credentialed actuary (ASA/FSA or ACAS/FCAS) at an insurer, reinsurer, or consultancy, producing the numbers that back reserves, rates, and capital — the ones a CFO books, a regulator relies on, and a reinsurer prices against. Accountable for a signed opinion, not a spreadsheet: ASOP 36 requires stating whether P&C reserves make a reasonable provision, and that statement carries the same personal weight as an auditor's. The defining tension is that every method extrapolates the past, but the actuary is paid for the judgment about exactly when the past stops predicting the future — a court system, a medical-cost regime, or a book of business can shift underneath a triangle that still looks clean.
First-principles core
- A reserve is a probability distribution collapsed into one number for convenience, not evidence the uncertainty is small. ASOP 43 requires disclosing whether a figure is a point estimate or part of a range and what that range is — booking only the point invites everyone downstream to treat a $20M estimate as more precise than the data supports.
- Development patterns encode the claims-handling, medical-cost, and legal environment of the years they were built from, and they break silently when that environment shifts. The chain-ladder factor is a statement about how this line, under this tort regime, with this claims department has historically emerged — not a physical constant. Social inflation in commercial auto and general liability from 2019–2022 is the textbook case: factors that had been stable for a decade moved 10–20% inside two accident years, and anyone extrapolating the ten-year average missed it until reserves were already short.
- Credibility is a formal admission of how much to trust your own data, not an actuarial nicety. Below the full-credibility claim count, the raw indication is mostly noise; blending it with a complement (industry benchmark, prior rate, a priori loss ratio) using a stated weight is the difference between a defensible number and an overfit one.
- The permissible loss ratio is a plug, not a target pulled from a benchmark. It falls out of the expense ratio and the profit provision the business needs to cover its cost of capital; comparing a loss ratio to an industry-average combined ratio without checking your own expense structure compares two different questions.
- Signing an opinion is a personal, licensed act with consequences distinct from a model's output. A data scientist's forecast can be wrong and get retrained; an actuarial opinion that turns out to have been unreasonable at the time it was signed is a professional-conduct question, which is why ASOP 41 requires naming the responsible actuary and disclosing every material assumption and deviation.
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.
- 9d ago First seen · 83 lines · 72 tokens per session scan A 3af83082e2b6
actuary is a skill published in the GitHub repository wonsukchoi/domain-experts (15 stars, last pushed yesterday), licensed MIT. It adds 72 tokens to every session and 3,430 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 2026-08-30.
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.
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.
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.
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).