← Back to Skills Marketplace
lofimichael

Longevity Assistant

by lofimichael · GitHub ↗ · v1.0.0
cross-platform ✓ Security Clean
2137
Downloads
3
Stars
6
Active Installs
1
Versions
Install in OpenClaw
/install longevity-assistant
Description
Evaluates longevity interventions using evidence tiers. Provides research evaluation framework and curated high-value insights on supplements, sleep, exercise, and protocols. Activate for anti-aging, healthspan, supplement evaluation, or research paper analysis.
README (SKILL.md)

Longevity Research Framework

Evidence-based longevity evaluation assistant. Teaches how to assess interventions using research methodology, not prescription. Provides curated non-obvious insights demonstrating the evaluation framework.

When to Activate

Trigger keywords: longevity, anti-aging, healthspan, lifespan, supplement evaluation, research paper analysis, evidence tier, biomarker interpretation, sleep optimization, exercise protocol, Bryan Johnson, Blueprint, mitochondria, autophagy, senolytics.

Evidence Tiers

Tier Definition Example
A Multiple RCTs, meta-analyses, consistent results Creatine for muscle
B Single RCT or large cohort, emerging human data Urolithin-A
C Mechanistic/animal studies, small human trials Most senolytics
D Anecdotal, theoretical, n=1 Novel peptides

Research Evaluation Framework

Study Design Hierarchy

  1. Systematic review / meta-analysis
  2. Randomized controlled trial (RCT)
  3. Cohort study (prospective > retrospective)
  4. Case-control study
  5. Case series / case reports
  6. Mechanistic / animal studies
  7. Expert opinion / theoretical

Assessment Checklist

  • Sample size: Adequately powered? (n>100 for most outcomes)
  • Duration: Appropriate for endpoint? (bone density needs years, not weeks)
  • Population: Relevant to you? (young athletes ≠ older adults)
  • Effect size: Clinically meaningful or just statistically significant?
  • Replication: Confirmed by independent groups?
  • Conflict of interest: Industry-funded? Disclosed relationships?

Red Flags

  • Single study with extraordinary claims
  • Surrogate endpoints only (biomarker change without clinical outcome)
  • Cherry-picked timepoints or subgroups
  • No control group or inadequate blinding
  • Massive effect sizes (>50% improvement = suspicious)
  • Published only in predatory journals
  • Funded entirely by supplement manufacturer
  • Authors selling the product

Alpha Discovery Framework

Use these patterns to identify non-obvious insights in longevity research:

Dosing Assumptions

  • Standard dose may not apply to all outcomes (tissue-specific thresholds)
  • "More is better" often has inverse U-curve (melatonin, antioxidants)
  • Saturation points differ by target (muscle vs. brain for creatine)

Timing & Context

  • Relative timing matters (cold exposure vs. training window)
  • Circadian timing affects efficacy (eating window, supplement timing)
  • Cycling may be required (adaptation, tolerance, microbiome shifts)

Form & Bioavailability

  • Same compound, different absorption (ethyl ester vs. triglyceride omega-3)
  • Conversion dependencies (ellagitannins → urolithin-A requires specific gut bacteria)
  • Cofactor requirements (fat-soluble vitamins need dietary fat)

Synergies & Antagonisms

  • Required pairings (D3 without K2 may cause harm)
  • Absorption competition (calcium and magnesium compete)
  • Timing conflicts (iron and coffee, cold and hypertrophy)

Population Specificity

  • Age-dependent responses (fasting + muscle loss in older adults)
  • Sex differences in metabolism
  • Genetic responders vs. non-responders (APOE and saturated fat)

Mechanism vs. Outcome

  • Plausible mechanism ≠ proven clinical benefit
  • Surrogate endpoints (biomarkers) ≠ real outcomes (mortality, function)
  • Animal doses rarely translate directly to humans

Example Alpha

The following examples demonstrate the discovery framework above. These are illustrative, not exhaustive—use the framework to evaluate new interventions.

Creatine: 15g for Cognitive Benefits

  • Common belief: 5g saturates muscle, same dose works for brain
  • Alpha: Serum creatine must rise high enough to cross blood-brain barrier and increase brain phosphocreatine. 5g saturates muscle but doesn't reliably raise brain levels.
  • Evidence: Multiple studies show cognitive benefits at 15-20g; 5g studies often null for cognition
  • Tier: B (emerging human data, mechanism understood)
  • Practical: Split 15g into 3x5g doses to avoid GI distress

Melatonin: 300mcg Outperforms 1mg+

  • Common belief: More melatonin = better sleep
  • Alpha: Body produces ~300mcg endogenously. Supraphysiological doses (1-10mg) cause next-day grogginess, may affect cognition long-term, and create dependency via receptor downregulation.
  • Evidence: Meta-analyses show 300mcg effective; higher doses don't improve outcomes
  • Tier: A (multiple meta-analyses)
  • Practical: Start at 300mcg; most commercial products are 10-30x too high

Urolithin-A: Mitophagy Without Pomegranate Roulette

  • Common belief: Eat pomegranates for mitochondrial health
  • Alpha: Urolithin-A (the active compound) requires gut bacteria conversion from ellagitannins. Only ~40% of people have the right microbiome. Direct supplementation bypasses this.
  • Evidence: PMC9133463, Timeline nutrition RCTs show mitophagy activation
  • Tier: B (human RCTs, mechanism validated)
  • Practical: 500-1000mg daily; one of few compounds with direct mitophagy evidence in humans

Sleep Timing > Sleep Duration

  • Common belief: Get 8 hours, timing doesn't matter
  • Alpha: Circadian rhythm governs 100+ physiological processes. Shifting sleep window by 2 hours causes more dysfunction than losing 1-2 hours of sleep. Late sleep (2am-10am) worse than short sleep (11pm-6am).
  • Evidence: Chronobiology research, shift-worker health outcomes
  • Tier: A (strong epidemiological + mechanistic)
  • Practical: Consistent bed/wake times matter more than duration optimization

Skin Damage: Cumulative and Irreversible

  • Common belief: Damage can be repaired with skincare products
  • Alpha: UV exposure causes cumulative DNA damage. Photoaging is largely irreversible. Prevention (sunscreen, clothing) has 100x ROI vs. treatment.
  • Evidence: Dermatology consensus, twin studies
  • Tier: A (decades of evidence)
  • Practical: Daily SPF 30+ on face/hands is highest-yield longevity intervention for appearance

Zone 2 Cardio: Mitochondrial Biogenesis

  • Common belief: HIIT is more efficient, Zone 2 is wasted time
  • Alpha: Zone 2 (can talk but not sing) specifically drives mitochondrial biogenesis and fat oxidation capacity. HIIT builds different adaptations. Both needed, but Zone 2 is undervalued.
  • Evidence: Exercise physiology, Inigo San Millan research
  • Tier: A (extensive mechanistic + performance data)
  • Practical: 3-4 hours/week Zone 2; most people go too hard and miss the adaptation

Cold Exposure: Timing Matters for Hypertrophy

  • Common belief: Cold exposure is universally beneficial
  • Alpha: Cold within 4 hours post-strength training blunts muscle protein synthesis and hypertrophy signaling. The inflammatory response you're suppressing is required for adaptation.
  • Evidence: Multiple mechanism studies, athletic performance research
  • Tier: B (consistent mechanism data, some human trials)
  • Practical: Cold exposure on rest days or 6+ hours after strength training

Berberine: Cycling Required

  • Common belief: Take daily like other supplements
  • Alpha: GI microbiome adapts to berberine, reducing effectiveness. Also, berberine's metformin-like effects may blunt some exercise adaptations.
  • Evidence: Clinical practice patterns, mechanism studies
  • Tier: B (clinical consensus, mechanism understood)
  • Practical: 4-6 weeks on, 2 weeks off; avoid on heavy training days

K2 (MK-7) + D3: Required Pairing

  • Common belief: Vitamin D alone is fine
  • Alpha: D3 increases calcium absorption. Without K2 to direct calcium to bones, it may deposit in arteries. K2 activates matrix-GLA protein and osteocalcin.
  • Evidence: Multiple RCTs, Rotterdam Study correlations
  • Tier: B (mechanistically clear, human outcome data emerging)
  • Practical: 100-200mcg MK-7 per 5000 IU D3; take together with fat

Omega-3: Form Affects Absorption 3x

  • Common belief: EPA/DHA amount is what matters
  • Alpha: Triglyceride and phospholipid forms have 3x better absorption than ethyl ester (most common in cheap supplements). Ethyl ester requires more fat for absorption.
  • Evidence: Bioavailability studies, head-to-head comparisons
  • Tier: A (well-established pharmacokinetics)
  • Practical: Pay more for triglyceride form or take ethyl ester with high-fat meal

Collagen: 15g+ for Joint Benefits

  • Common belief: Small amounts help skin/joints
  • Alpha: Studies showing joint benefits used 10-15g doses. Lower doses may help skin hydration but don't move the needle on joint tissue synthesis.
  • Evidence: Joint-specific RCTs used higher doses than skin studies
  • Tier: B (human RCTs at effective dose)
  • Practical: 15g+ if targeting joints; 5g may suffice for skin only

Fasting: Protein Timing Beats Duration

  • Common belief: Longer fasts are better
  • Alpha: Muscle protein synthesis (MPS) is pulsatile. Extending fasts beyond 16-18h risks muscle catabolism, especially over age 40. Early time-restricted eating (eating earlier in day) outperforms late eating windows.
  • Evidence: MPS research, circadian metabolism studies
  • Tier: B (mechanism clear, human data supportive)
  • Practical: 16:8 with eating window 8am-4pm beats 20:4 with window 2pm-6pm

Safety Principles

  1. Physician consultation: Required for existing conditions, medications, or symptoms
  2. One variable at a time: Introduce supplements individually, 1-2 week gaps
  3. Start at 50% dose: Titrate up based on response
  4. Stop before surgery: Most supplements stopped 1-2 weeks pre-surgery
  5. Watch for interactions: Blood thinners, thyroid meds, and blood pressure meds have many supplement interactions

This skill does not diagnose, treat, or prescribe. All information is educational.


Extended Capabilities

When tools are available:

  • Web search: Query PubMed for recent studies, verify safety alerts
  • File reading: Analyze uploaded lab results or research papers
  • Calculation: HOMA-IR, dosing by body weight, cost-per-dose comparisons

Example queries for research:

  • "[compound] site:pubmed.gov RCT 2024 OR 2025"
  • "[supplement] meta-analysis systematic review"

Guidelines

Always

  • Cite evidence tiers for recommendations
  • Distinguish mechanism (plausible) from outcome (proven)
  • Acknowledge uncertainty and individual variation
  • Recommend professional consultation for medical concerns

Never

  • Diagnose or prescribe
  • Overstate evidence quality (C-tier is not "proven")
  • Ignore potential interactions
  • Guarantee outcomes
Usage Guidance
This skill appears internally consistent and low-risk technically because it is instruction-only and requests no system access or credentials. Things to consider before installing or relying on it: - Provenance: the skill's source/homepage are unknown; verify any important claims by checking primary literature and trusted sources. - Medical caution: although the SKILL.md frames itself as educational, it contains concrete dosing examples and practical recommendations — do not treat these as medical advice; consult a qualified clinician before acting. - Verify citations: when the skill cites studies or specific evidence (some examples reference RCTs or PMC IDs), confirm the studies and their quality yourself rather than relying solely on the summary. - Privacy/security: the skill does not request credentials or perform network installs, so it cannot exfiltrate secrets by design; however, content could still be harmful if followed blindly. If you want greater assurance, ask the publisher for source attribution or references for the specific claims and dosing examples before relying on them.
Capability Analysis
Type: OpenClaw Skill Name: longevity-assistant Version: 1.0.0 The skill bundle is benign. The `SKILL.md` file provides extensive educational content and clear guidelines for the AI agent's behavior, focusing on longevity research and evaluation. While it mentions 'Web search' and 'File reading' capabilities, these are explicitly framed within the context of the skill's stated purpose (e.g., querying PubMed for research, analyzing uploaded lab results or research papers). There is no evidence of intent to exfiltrate data, execute arbitrary commands, or perform any other malicious actions, nor are there any prompt injection attempts designed to subvert the agent's intended role for harmful purposes.
Capability Assessment
Purpose & Capability
The SKILL.md content matches the name and description: it provides an evidence-tier framework, triggers, and illustrative examples for longevity research. There are no unexpected binaries, environment variables, or config paths required that would be inconsistent with an educational/evaluation assistant. Note: the skill's source/homepage are not provided, but that affects provenance rather than technical coherence.
Instruction Scope
Instructions are limited to describing evaluation heuristics, tiers, and illustrative examples. The document does not direct the agent to read local files, access environment variables, or send data to external endpoints. One notable point: the SKILL.md contains specific dosing examples and practical recommendations (e.g., creatine 15g, melatonin 300mcg, urolithin-A doses). While the file states it is educational and not a prescription, those concrete dosing items are actionable content — this is a content-safety/medical-consideration note, not a technical inconsistency.
Install Mechanism
There is no install specification and no code files. As an instruction-only skill it does not write code or download artifacts during install, which is the lowest-risk install profile.
Credentials
The skill declares no required environment variables, credentials, or config paths and the instructions do not reference any secrets. There is no disproportionate credential request relative to the stated purpose.
Persistence & Privilege
always is false and the skill does not request persistent or elevated privileges. It does not instruct changes to other skills or system-wide settings.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install longevity-assistant
  3. After installation, invoke the skill by name or use /longevity-assistant
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Longevity Assistant 1.0.0 - Initial release providing a detailed research evaluation framework for longevity interventions. - Introduces a 4-tier evidence rating system (A–D) for assessing supplements, protocols, and interventions. - Includes a practical checklist and red flags for research paper evaluation. - Offers the "Alpha Discovery Framework" for finding high-leverage, non-obvious insights about dosing, timing, synergy, bioavailability, and more. - Contains curated evidence-based examples demonstrating the framework across sleep, exercise, supplements, and anti-aging strategies. - Designed for use in anti-aging, healthspan optimization, supplement evaluation, and research paper analysis. source: https://github.com/lofimichael/longevity-skill
Metadata
Slug longevity-assistant
Version 1.0.0
License
All-time Installs 6
Active Installs 6
Total Versions 1
Frequently Asked Questions

What is Longevity Assistant?

Evaluates longevity interventions using evidence tiers. Provides research evaluation framework and curated high-value insights on supplements, sleep, exercise, and protocols. Activate for anti-aging, healthspan, supplement evaluation, or research paper analysis. It is an AI Agent Skill for Claude Code / OpenClaw, with 2137 downloads so far.

How do I install Longevity Assistant?

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

Is Longevity Assistant free?

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

Which platforms does Longevity Assistant support?

Longevity Assistant is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Longevity Assistant?

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

💬 Comments