← Back to Skills Marketplace
ivangdavila

Health Insurance

by Iván · GitHub ↗ · v1.0.0
linuxdarwinwin32 ✓ Security Clean
341
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install health-insurance
Description
Compare health insurance plans, estimate total yearly costs, and choose coverage that fits medical usage, prescriptions, and financial risk.
README (SKILL.md)

Setup

On first use, read setup.md for integration guidelines and memory initialization.

When to Use

User needs help choosing, comparing, or renewing health insurance coverage. Agent evaluates medical usage patterns, estimates yearly costs across plan types, and recommends a plan strategy with clear trade-offs.

Architecture

Memory lives in ~/health-insurance/. See memory-template.md for structure.

~/health-insurance/
├── memory.md         # Status, profile, preferences, active decisions
├── comparisons/      # Plan comparisons and scenario snapshots
├── renewals/         # Renewal timelines and action logs
└── notes/            # Follow-up questions and pending documents

Quick Reference

Topic File
Setup process setup.md
Memory template memory-template.md
Coverage framework coverage-framework.md
Annual cost modeling cost-model.md
Comparison checklist comparison-checklist.md
Enrollment and renewal playbook enrollment-playbook.md

Core Rules

1. Lock Decision Context First

Before comparing plans, clarify:

  • Coverage target: individual, couple, or family
  • Source: employer plan, public marketplace, private broker, or government program
  • Geography and provider access requirements
  • Hard constraints: budget ceiling, medication continuity, renewal deadline

2. Build a Real Utilization Profile

Estimate expected care load before discussing premiums:

  • Routine care frequency (primary care, specialist, urgent care)
  • Ongoing prescriptions and refill cadence
  • Known procedures, therapies, or recurring diagnostics
  • Worst-case risk profile for unexpected events

3. Compare Plan Mechanics Before Price

Always evaluate these mechanics before deciding on monthly premium:

  • Network fit for current clinicians and facilities
  • Deductible, out-of-pocket max, and coinsurance structure
  • Copay design by care type (primary, specialist, urgent, emergency)
  • Referral and prior-authorization friction for expected treatments
  • Prescription formulary coverage for required medications

4. Model Yearly Cost With Scenarios

Use cost-model.md to calculate low, expected, and high-use annual totals. Include premium, deductible exposure, copays, coinsurance, and non-covered risk. Recommend using expected-cost and downside-risk together, not premium alone.

5. Protect Against Coverage Failure Modes

Run a risk check before final recommendation:

  • Out-of-network emergency and balance-billing exposure
  • Drug tier surprises and step-therapy limitations
  • Referral bottlenecks that delay care
  • High deductible plans that look cheap but shift excessive risk

6. Execute Enrollment With Evidence

Use enrollment-playbook.md to define exact actions, deadlines, and proof artifacts. Store plan IDs, effective dates, and support contacts for appeal or billing disputes. Never claim enrollment is complete without confirmation evidence.

7. Persist Data Only With Explicit Approval

Before writing to ~/health-insurance/memory.md, ask for explicit confirmation. Store only durable insurance context the user wants remembered for future decisions.

Common Traps

  • Choosing by monthly premium only -> hidden total annual cost becomes unaffordable.
  • Ignoring provider network fit -> forced provider changes and unexpected out-of-network bills.
  • Skipping formulary checks -> medication cost spikes after enrollment.
  • Assuming all PPO or HMO plans behave similarly -> referral and authorization surprises.
  • Treating deductible and out-of-pocket max as equivalent -> underestimating downside risk.
  • Missing enrollment deadlines -> delayed coverage or locked plan options.

External Endpoints

This skill makes NO external network requests.

Endpoint Data Sent Purpose
None None N/A

No data is sent externally.

Security & Privacy

Data that leaves your machine:

  • Nothing. This skill is instruction-only and local by default.

Data stored locally:

  • Insurance profile and comparison context explicitly approved by the user.
  • Stored in ~/health-insurance/memory.md.

This skill does NOT:

  • Access insurer or broker APIs automatically.
  • Submit enrollment forms or claims without user direction.
  • Read files outside ~/health-insurance/ for storage.
  • Write memory without explicit user confirmation.
  • Modify its own core instructions or auxiliary files.

Trust

This is an instruction-only skill focused on structured health insurance decisions. No credentials are required and no external service access is needed.

Related Skills

Install with clawhub install \x3Cslug> if user confirms:

  • health — health planning context that informs insurance priorities
  • doctor — provider interaction planning and visit preparation
  • compare — structured side-by-side decision frameworks
  • money — budgeting and cash-flow planning for premium and out-of-pocket costs

Feedback

  • If useful: clawhub star health-insurance
  • Stay updated: clawhub sync
Usage Guidance
This skill is instruction-only and appears coherent, but consider these practical checks before enabling it: - Confirm you are comfortable letting the agent create ~/health-insurance/ and write memory.md; the skill promises to ask for explicit approval before writing. If you decline, it should not persist data. - Review the contents of any created files (memory.md, comparisons, notes) before storing sensitive data — the skill explicitly recommends not storing policy numbers, government IDs, or payment credentials. - The skill declares it makes no external network requests, but that relies on the agent runtime and policies. If you run agents that can access the network, ensure your environment/network controls match your privacy needs. - The setup uses simple shell commands (mkdir/touch/chmod). If you run an agent with capability to execute system commands, be aware those same privileges could write files elsewhere; rely on the agent platform's sandboxing and consent prompts. If these points are acceptable, the skill is coherent with its stated purpose and low-risk in isolation. If you need stricter guarantees, ask the developer for an explicit promise/logging of write actions or use the skill in a restricted environment.
Capability Analysis
Type: OpenClaw Skill Name: health-insurance Version: 1.0.0 The skill bundle is classified as benign. The `SKILL.md` explicitly states that it makes no external network requests and does not exfiltrate data. While `setup.md` contains shell commands (`mkdir`, `touch`, `chmod`) for local file system operations, these are explicitly conditioned on the agent asking for and receiving 'explicit approval' from the user before execution. The commands themselves are for creating a dedicated local directory (`~/health-insurance/`) with restrictive permissions (`chmod 700`, `chmod 600`), which aligns with the skill's stated purpose of local memory management and demonstrates good security practices. No evidence of prompt injection, malicious execution, persistence, or obfuscation was found.
Capability Assessment
Purpose & Capability
Name, description, and all provided files (setup, templates, checklists, cost model, playbook) align with a tool for comparing health plans and modeling annual costs. Requested filesystem locations (~/health-insurance/) are coherent for persisting user-approved decision memory.
Instruction Scope
Runtime instructions are limited to local operations: reading supplied docs, building a local memory folder, and running local checks. The skill explicitly states it makes no external network requests and requires explicit user approval before writing memory. Setup includes simple shell filesystem commands (mkdir, touch, chmod) which are appropriate for initializing local storage.
Install Mechanism
No install spec or code is provided — this is instruction-only which minimizes install-time risk and nothing is downloaded or executed from external sources by the skill itself.
Credentials
The skill requests no environment variables, credentials, or config paths outside its declared local memory area. It also documents not storing sensitive identifiers or payment credentials.
Persistence & Privilege
always:false (no forced persistent inclusion). The only persistent behavior described is optional local memory creation under explicit user approval, with recommended restrictive file permissions (chmod 700/600). The skill does not claim to modify other skills or global agent settings.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install health-insurance
  3. After installation, invoke the skill by name or use /health-insurance
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release with plan comparison workflows, annual cost modeling, enrollment timing guidance, and local memory for recurring insurance decisions.
Metadata
Slug health-insurance
Version 1.0.0
License
All-time Installs 1
Active Installs 1
Total Versions 1
Frequently Asked Questions

What is Health Insurance?

Compare health insurance plans, estimate total yearly costs, and choose coverage that fits medical usage, prescriptions, and financial risk. It is an AI Agent Skill for Claude Code / OpenClaw, with 341 downloads so far.

How do I install Health Insurance?

Run "/install health-insurance" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Health Insurance free?

Yes, Health Insurance is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Health Insurance support?

Health Insurance is cross-platform and runs anywhere OpenClaw / Claude Code is available (linux, darwin, win32).

Who created Health Insurance?

It is built and maintained by Iván (@ivangdavila); the current version is v1.0.0.

💬 Comments