selector-ultimates-ai-framework

selector-ultimates-ai-framework is an agent for Claude Code from cas-team-analyst/team-analyst. It costs 98 tokens per session (3,800 once invoked), scanned B, original, MIT.

An AI framework for choosing ultimate insurance losses and claim counts by accident year. Ultimate losses are the estimated final amounts, and accident year groups claims by the year the accident happened.

In plain words
What is it for?
Use it to select loss and count ultimates from method results, triangle diagnostics, exposure data, and prior selections.
Why use it?
It applies documented judgment to choose between actuarial methods and paid or incurred data instead of leaving the final selection to an unexplained guess. It requires context files supplied by a parent agent.

Agent for Claude Code

Written for Claude Code: user-invocable in frontmatter.

Part of the team-analyst plugin — 7 skills, 6 agents shipped together

Good fit Use it to select loss and count ultimates from method results, triangle diagnostics, exposure data, and prior selections.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/cas-team-analyst/team-analyst/selector-ultimates-ai-framework
Install

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.

Clone the repo
git clone --depth 1 https://github.com/cas-team-analyst/team-analyst

Made for: Claude Code.

Or install team-analyst, the plugin that ships this one along with the rest of its 7 skills, 6 agents.

Wrote 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.

agentmods badge for selector-ultimates-ai-framework

README.md
[![agentmods](https://agentmods.dev/badge/agents/cas-team-analyst/team-analyst/selector-ultimates-ai-framework.svg)](https://agentmods.dev/agents/cas-team-analyst/team-analyst/selector-ultimates-ai-framework)
Your own site
<a href="https://agentmods.dev/agents/cas-team-analyst/team-analyst/selector-ultimates-ai-framework"><img src="https://agentmods.dev/badge/agents/cas-team-analyst/team-analyst/selector-ultimates-ai-framework.svg" alt="Measured on agentmods" height="20"></a>
Per session 98 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,800 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00098 $0.03800
Opus 5 $0.00049 $0.01900
Sonnet 5 $0.00020 $0.00760
Haiku 4.5 $0.00010 $0.00380

Measured 7d ago against content hash 4e728101c5e4, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade B, and why

selector-ultimates-ai-framework scanned grade B with 1 finding 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 7d 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.

Asks the agent to reveal its instructionsmediumSystem prompt leakage

Directions to print, repeat or translate the system prompt extract configuration the operator did not intend to expose.

8. Write a JSON file for that category with full reasoning, per Output Instructions below
skills/reserving-analysis/agents/selector-ultimates-ai-framework.agent.md · 233 lines

How it starts

The opening of the file, as written. The whole thing — 233 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an expert P&C actuarial analyst selecting ultimate losses and counts by accident year from a set of method indications. You read method outputs, triangle diagnostics, exposure data, and prior selections provided as text, apply the framework below, and return JSON selections for Loss and Count categories.

You do not write or execute a script to apply this framework. The selection criteria below have too many interacting, judgment-laden conditions to encode reliably in code. Work through them yourself, by reasoning, for each accident year.

IMPORTANT: You are making TWO selections per accident year:

  1. One Loss ultimate (choosing between Incurred Loss and Paid Loss indications)
  2. One Count ultimate (choosing between Reported Count and Closed Count indications)

The parent agent will provide you with two context file paths: one for Loss, one for Count.

Your first step: The parent agent will pass you a list of context markdown file paths (e.g., selections/ultimates-context-loss.md, selections/ultimates-context-count.md). These are your primary data sources. Do not rely on Ultimates.xlsx as primary input because formula cells may not be evaluated in headless runs. Do not read all of them now — process one category at a time following the read/write loop in the Task section below.

Task

For each category (Loss and Count):

  1. Read the category's context file (e.g., selections/ultimates-context-loss.md) - only one at a time.
  2. Review all available method indications for both measures in the category (e.g., Incurred Loss and Paid Loss for the Loss category)
  3. Work through the Selection Criteria below in order, phase by phase
  4. Apply any Situational Modifiers that fit this line and period
  5. Choose ONE ultimate per accident year - selecting the measure (Incurred vs Paid, or Reported vs Closed) and method combination that best represents the expected ultimate based on maturity, data quality, and diagnostics
  6. Run the Cross-Cutting Checks once all periods in the category have a selection
  7. Always return a selection for every period provided, including the oldest (tail-exposed) year
  8. Write a JSON file for that category with full reasoning, per Output Instructions below
  9. Move to the next category.

Selection Philosophy: For each accident year, you are choosing the SINGLE BEST ultimate estimate, not weighting across measures. Consider: Which measure (Incurred vs Paid, Reported vs Closed) is more credible at this maturity? Which methods are most appropriate for that measure? What is the final ultimate value?


Read the full file on GitHub · 233 lines

Changes

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.

  1. 7d ago Changed · +94 lines 4e728101c5e4
  2. 8d ago First seen · 139 lines · 98 tokens per session scan B fdd33dbcd206

Subscribe to this mod's changes

selector-ultimates-ai-framework is an agent published in the GitHub repository cas-team-analyst/team-analyst (11 stars, last pushed 4d ago), licensed MIT. It adds 98 tokens to every session and 3,800 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it B with 1 finding (asks the agent to reveal its instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other agents, from other repositories

chrono

Temporal Pattern Expert analyzing time-of-day, day-of-week, and seasonality.

emerzon/mtdata-mcp · 17 tokens

financial-integrity-auditor

Audits ONE completed CodeOps phase diff for monetary-correctness defects — idempotency of money-moving operations, duplicate-submission and double-spend windows, rounding and precision, atomicity and rollback on partial failure, reconciliation, audit-trail completeness, negative and overflow amounts, currency and unit…

blendsdk/claude-codeops · 142 tokens

token-economics-designer

Token economics and tier design specialist. Use when designing pricing models, access tiers, or token distribution strategies.

Matt-Dionis/claude-code-configs · 28 tokens

soleur-finance-budget-analyst

Use this agent when you need to create budget plans, analyze spending allocation, model burn rate scenarios, or review budget-to-actual variance. Use ops-advisor for expense tracking and vendor cost research; use this agent for budget planning and allocation analysis. Use cfo for cross-cutting financial strategy.

jikig-ai/soleur · 69 tokens

ic-challenger

The toughest person on the investment committee with 30 years of CRE experience spanning three full cycles. Challenges every assumption with data and forces analysts to defend their work to the highest standard. Produces structured challenge memos that systematically stress-test investment theses. Deploy this agent…

mariourquia/cre-skills-plugin · 85 tokens

pm-business-analyst

Business analysis agent for business cases, market sizing, and financial modeling. Invoke when users need to assess investment feasibility, calculate TAM/SAM/SOM, model pricing strategies, analyze unit economics (LTV, CAC, NRR), or build financial projections. Trigger keywords: business case, financial model, market…

geekatron/jerry · 107 tokens