Skip to main content
skill CX Codex

risks — Claude Code skill: health risk forecasts and setup

This Claude Code skill estimates a person’s main health risks from their recorded state and evidence. It runs when PLSI requests the future part of a life trajectory or a user asks about their risks.

MIT ★ 93 updated 2026-11-15

Install into Claude Code · Skills CLI (Vercel) All methods and platforms ↓
npx skills add Insilico-org/plsi --skill risks -a claude-code
Open on GitHub

What risks does

The skill selects relevant events and diagnoses from personal history, family history, and current state, then assigns each a level and time window. For a numeric estimate, it looks for a risk equation, a matching base-rate table, or a verified cohort paper. If none fits, it gives a level with a hypothesis basis instead of a probability. No source means no probability.

For equation-based estimates and what-if inputs, it calls plsi calc derive rather than doing the arithmetic itself. It reads state, care context, due items, and plans through plsi commands; PLSI must provide a processor/history result or a request about risks. The output is one result containing events, evidence links, and the top modifiable risk. An adopted plan adds two trajectory courses.

What it helps with

  • rank future health events from personal and family history
  • estimate absolute risk using a relevant source
  • distinguish a current check from a future event
  • compare a forecast with an adopted plan

Example prompts for the agent

  • Assess my main health risks from my current state and history.
  • Build the future part of my life trajectory, with risk windows and sources.
  • Compare the forecast with my adopted plan and show both trajectory courses.

How to install risks

1Skills CLI (Vercel)

npx skills add Insilico-org/plsi --skill risks -a claude-code

2Manually: put the files in place

~/.claude/skills/risks/SKILL.md

Copy the whole skill folder; for one project use .claude/skills/risks/

Step by step, with commands for each platform: How to install skills in Claude Code: five ways

Questions about risks

How do I install risks in Claude Code?

Run npx skills add Insilico-org/plsi --skill risks -a claude-code. The skill file is placed at ~/.claude/skills/risks/SKILL.md.

What does risks do?

It selects future events that matter to the person and estimates their risks using equations, base-rate tables, or verified cohort papers. Each event gets a time window and a source.

How does risks calculate probabilities?

It seeks a suitable risk equation, a base-rate table for a matching group, or a verified cohort paper. When no source fits, it reports a level without a numeric probability.

Source text

SKILL.md · MIT
---
name: risks
description: Use as PLSI Nostradamus when PLSI asks for the future part of the life arrow or for the user's main health risks - pick the 3-7 events that matter most, each with a window and a sourced absolute risk from risk equations, base-rate tables or cohort papers.
---

# PLSI risks

## When to use

- PLSI hands over PLSI Data Processor's `processor/history` result for the life arrow (`orchestrator/arrow`).
- The user asks "what are my main risks" or "what is likely ahead for me".

## Steps

1. **Read the state** cited in the request, the care and personal context, and what is already planned:

   ```bash
   plsi state show --instance <instance> --map <map_id>
   plsi state show --instance <instance> --map care_context
   plsi due --instance <instance>
   plsi plan list --instance <instance>
   ```

2. **Pick 3-7 events** from the user's own state, gaps, history (resolved conditions too: `plsi events list
   --instance <instance> --kind condition`) and family history: the ones that are likely within
   the window, or severe and modifiable. No fixed list. Each is an event or diagnosis that can be dated, not a
   process ("harm building", "features building").

3. **Source each probability**, in this order; stop at the first that fits:
   - a risk equation in a published state; if no map has it, put a `processor/add-score` request in your `result`;
   - a `base_rate` table for the person's age, sex, country and period (`plsi refs list`), cited with its year;
   - a verified cohort paper:

     ```bash
     plsi refs list
     plsi evidence search --source pubmed --query "<event> absolute risk <risk factors> cohort"
     plsi evidence verify --kind pmid --id <pmid>
     ```

     Reuse the evidence library and verify only what you cite (`shared/team`, Evidence speed).

   - otherwise no number: a level only, `"basis": "hypothesis_only"`.

   The source must match what is recorded: when symptoms or findings raise the risk (witnessed breathing pauses, a
   raised marker), use a rate for people with those findings, never the general-population rate.

   Risk from an equation or with what-if inputs comes from `plsi calc derive`, never by hand.

4. **Level** each event: `high` (likely within the window, or severe and modifiable), `watch` (raised vs the matching
   group), `info`. Add scheduled re-measurements and recorded plan predictions (step 1) as `info` events.

5. **Return ONE `result`**: per event the window, label, level, probability (or none), basis and link; the top
   modifiable risk named first. With an adopted plan, add `nostradamus/trajectory`'s two courses.

## Rules

- No probability without a source; relative risks are turned into absolute ones only against a cited base rate.
- Name no intervention as a cure; changes of course come only from plans that passed `plsi screen`.
- Rank by what matters to this person, not by how alarming an event sounds.
- A condition whose signs are present now (snoring with witnessed pauses and daytime sleepiness) is a "check now"
  item, not a future event.
- A care step (a visit, a test) carries a time frame from a cited guideline; never a later one.

© Insilico-org · MIT · Insilico-org/plsi

Similar skills

All skills for Science and research →

From the same set85

The whole set →
clinics This Claude Code skill finds clinics and labs for tests or check-ups. It ranks nearby options by services, reviews, distance, and budget. CX Codex +1 ★ 93repo stars trials This Claude Code skill finds clinical trials using a person’s health context, goals, and location. It activates when PLSI is asked about joining a trial or has no suitable standard option for a goal. CX Codex +1 ★ 93repo stars devices This Claude Code skill compares at-home tests and devices by validation evidence, cost, and export options. It runs when PLSI needs a device recommendation or home measurements to fill map gaps. CX Codex +1 ★ 93repo stars hevy This Claude Code skill connects PLSI to the Hevy workout log and shares training data with Coach. It runs for connection setup, checks, repairs, removal, or workout-history requests. CX Codex +1 ★ 93repo stars explain A Claude Code skill for turning research findings into competing, testable explanations. It activates when plsi research next selects explain. CX Codex +1 ★ 93repo stars diet This Claude Code skill builds a nutrition plan around health goals, medical conditions, allergies, and food preferences. It activates for requests about diet, weight, or eating with high cholesterol or glucose. CX Codex +1 ★ 93repo stars history The history Claude Code skill records health events from a personal story, a conversation, or another bot’s notes. It updates relevant PLSI maps and returns past events alongside the current state. CX Codex +1 ★ 93repo stars substances The substances Claude Code skill builds plans to quit or reduce smoking, vaping, or alcohol use. It checks for risks from stopping without supervision and sets a stepwise schedule. CX Codex +1 ★ 93repo stars

Project sponsors

Companies that keep this analytics open