healthcare-agent-instruction-letter

Drafts a non-binding healthcare values/instruction letter for a designated health care agent to guide substituted-judgment decisions during incapacity. Use when a client needs a bedside document that harmonizes with a health care power of attorney, healthcare proxy, or advance directive and captures quality-of-life thresholds, treatment preferences, and conflict guidance. Trigger on requests for "instruction letter," "health care agent letter," "values letter," "legacy letter," "healthcare proxy," or "DPOA health care."

11 stars

Best use case

healthcare-agent-instruction-letter is best used when you need a repeatable AI agent workflow instead of a one-off prompt.

Drafts a non-binding healthcare values/instruction letter for a designated health care agent to guide substituted-judgment decisions during incapacity. Use when a client needs a bedside document that harmonizes with a health care power of attorney, healthcare proxy, or advance directive and captures quality-of-life thresholds, treatment preferences, and conflict guidance. Trigger on requests for "instruction letter," "health care agent letter," "values letter," "legacy letter," "healthcare proxy," or "DPOA health care."

Teams using healthcare-agent-instruction-letter should expect a more consistent output, faster repeated execution, less prompt rewriting.

When to use this skill

  • You want a reusable workflow that can be run more than once with consistent structure.

When not to use this skill

  • You only need a quick one-off answer and do not need a reusable workflow.
  • You cannot install or maintain the underlying files, dependencies, or repository context.

Installation

Claude Code / Cursor / Codex

$curl -o ~/.claude/skills/healthcare-agent-instruction-letter/SKILL.md --create-dirs "https://raw.githubusercontent.com/CaseMark/skills/main/skills/legal/healthcare-agent-instruction-letter/SKILL.md"

Manual Installation

  1. Download SKILL.md from GitHub
  2. Place it in .claude/skills/healthcare-agent-instruction-letter/SKILL.md inside your project
  3. Restart your AI agent — it will auto-discover the skill

How healthcare-agent-instruction-letter Compares

Feature / Agenthealthcare-agent-instruction-letterStandard Approach
Platform SupportNot specifiedLimited / Varies
Context Awareness High Baseline
Installation ComplexityUnknownN/A

Frequently Asked Questions

What does this skill do?

Drafts a non-binding healthcare values/instruction letter for a designated health care agent to guide substituted-judgment decisions during incapacity. Use when a client needs a bedside document that harmonizes with a health care power of attorney, healthcare proxy, or advance directive and captures quality-of-life thresholds, treatment preferences, and conflict guidance. Trigger on requests for "instruction letter," "health care agent letter," "values letter," "legacy letter," "healthcare proxy," or "DPOA health care."

Where can I find the source code?

You can find the source code on GitHub using the link provided at the top of the page.

SKILL.md Source

# Healthcare Agent Instruction Letter

Drafts a scenario-based, non-binding values letter in the client's voice to empower the health care agent and reduce ambiguity at the bedside. Output supplements—never replaces—operative legal directives.

## Quick Start

1. Confirm client identity, preferred name, and jurisdiction.
2. Collect executed documents (name/date/type): health care POA or proxy, advance directive, POLST/MOLST, HIPAA release.
3. Identify primary agent and alternates (full names, roles, contact preferences).
4. Capture core values: quality-of-life thresholds, cognition-loss tolerance, dependence, pain, institutional care.
5. Capture clinical preferences: CPR, ventilation, artificial nutrition/hydration, antibiotics, surgery, time-limited trials, hospice, comfort care.
6. Capture spiritual/cultural commitments affecting care decisions.
7. Identify conflict vectors and family communication plan.
8. Capture tone preference, distribution timing, and document-storage locations.
9. Never invent names, dates, diagnoses, or beliefs—ask for missing facts and use `[placeholder]`.

## Core Workflow

| Phase | Action | Output |
|-------|--------|--------|
| 1 | Validate non-binding status | Opening disclaimer: guidance only, not a substitute for operative directives |
| 2 | Harmonize documents | List governing document titles/dates; note instruction hierarchy |
| 3 | Elicit decision framework | Big-Three thresholds: cognition, independence, suffering |
| 4 | Draft scenario logic | Treatment sections with conditional "if/when…then…" examples |
| 5 | Add faith/culture mechanics | Practical care requests tied to beliefs, without legal overstatement |
| 6 | Address conflict protocol | Disagreement scripts, consult triggers, clinician communication |
| 7 | Add logistics | Distribution plan: who receives, when, storage locations |
| 8 | Run quality controls | Mismatch flags, ambiguity reduction, attorney-review gate |

### Writing Constraints

- Client voice, plain language, short headings, 1–3 pages.
- Conditional framing throughout: "if/when…then…" with scenario examples.
- No medical orders or statutory language belonging in POLST/MOLST/clinic forms.
- Every section must be usable under stress by one reader in minutes.

## Letter Template

Fill only known facts; use `[placeholder]` where missing.

```text
Date: [date]
To: [Primary Agent], [Alternate(s)]
Re: My Values and Medical Decision Guidance

I wrote this letter to guide my agent if I cannot speak for myself.
I have executed [document names/dates]; this letter supplements and does not replace them.

1) Decision-Makers
- [Primary Agent], [Alternate Agent]
- Use substituted judgment: do what I would want, not what feels least difficult.

2) What Matters Most
- Good-day definition:
- Non-negotiable values/limits:
- Cognitive/function thresholds:
- Pain, comfort, dignity standards:

3) Treatment Preferences
- CPR / intensive care:
- Ventilation / breathing support:
- Feeding / hydration:
- Dialysis / infection treatment:
- Surgery / rehab vs. noninvasive care:
- Time-limited trials:
- Pain control / sedation:
- Preferred care setting:

4) Spiritual, Cultural, and Personal Preferences
- Practices/rituals:
- Clergy or contacts:
- Religious limits or accommodations:

5) Conflict and Communication
- Who should be included:
- Who may contest:
- Protocol: ethics/palliative consults, second opinions, family meeting trigger:

6) Closing
- Permission grant:
- Document storage locations:
- Distribution instructions:

Signature:
```

## Post-Draft Checks

- [ ] Non-binding disclaimer appears in opening and closing.
- [ ] No contradiction with operative legal documents; discrepancies flagged separately.
- [ ] No invented facts; `[placeholder]` tags remain for missing data.
- [ ] Treatment guidance is scenario-based, not slogans.
- [ ] Conflict instructions protect agent authority consistent with legal appointment.
- [ ] No absolute commands implying standalone legal effect.
- [ ] Attorney-review gate and confidentiality note included.

## Jurisdictional Guardrails

| Topic | Rule |
|-------|------|
| Terminology | Match client documents: "health care proxy," "medical power of attorney," "advance health care directive," etc. |
| POLST/MOLST | Do not draft as an order form; recommend separate clinical workflow [VERIFY]. |
| Heightened-evidence jurisdictions | Use heightened specificity where clear-and-convincing evidence may be required for life-sustaining-treatment disputes (e.g., *In re Storar*, 52 N.Y.2d 363 (1981)) [VERIFY]. |
| MAID | Do not imply the agent can request aid-in-dying on the principal's behalf; separate statutory process [VERIFY]. |
| Default surrogacy | Reference only client-selected legal chain; do not summarize statutes without attorney verification. |

## Pitfalls

**Do:**
- Keep language operational: who, when, what threshold, what action.
- Preserve client's voice and style.
- Use `[placeholder]` for all absent details.
- Require attorney review before final use.

**Don't:**
- Treat this as an enforceable legal order or substitute advance directive.
- Include medical advice or treatment guarantees.
- Create contradictions with signed legal documents.
- Use inflammatory family language that could weaponize disputes.
- Advise distribution to clinicians without confirming client disclosure preferences.

Attorney review is mandatory. Draft output is informational assistance only and does not constitute legal advice.

Related Skills

We are still matching the closest adjacent skills for this page. In the meantime, continue through the full directory.