← Back to Skills Marketplace
agimodel

Fit

by AGImodel · GitHub ↗ · v1.0.0
cross-platform ✓ Security Clean
304
Downloads
0
Stars
1
Active Installs
1
Versions
Install in OpenClaw
/install fit
Description
Your personal fitness operating system. Not a workout plan. A complete system that figures out what your body actually needs, builds training around your rea...
README (SKILL.md)

Fit

What Most Fitness Advice Gets Wrong

Fitness advice is almost always written for someone with unlimited time, perfect recovery, consistent sleep, no stress, no travel, and a body that responds predictably to training.

That person does not exist.

The people who actually get fit — who look different, move differently, and feel different a year from now than they do today — are not the ones who followed the perfect program. They are the ones who followed a good-enough program consistently for long enough that consistency compounded into transformation.

This skill is built for real people with real constraints.


The System

FITNESS_OS = {
  "assessment": {
    "current_state": ["Current weight and body composition if known",
                      "Current training frequency and type",
                      "Movement limitations or injuries",
                      "Energy levels across the day",
                      "Sleep quality and quantity"],
    "goal_clarity":  {
      "aesthetic":     "Body composition change — fat loss, muscle gain, or both",
      "performance":   "Specific metric — run 5K, deadlift bodyweight, 10 pullups",
      "health":        "Blood markers, blood pressure, longevity, pain reduction",
      "functional":    "Move better, carry more, hurt less"
    },
    "constraint_map": ["Days per week available for training",
                       "Minutes per session realistic",
                       "Equipment available",
                       "Injuries or restrictions",
                       "Travel frequency"]
  }
}

Training Architecture

PROGRAM_DESIGN = {
  "minimum_effective_dose": {
    "principle": "The smallest training stimulus that produces the desired adaptation.
                  More is not better. Enough is better.",
    "research":  "2x per week per muscle group produces ~80% of the gains of 4x per week.
                  3x per week per muscle group captures nearly all available adaptation.",
    "implication": "A 3-day full-body program done consistently beats a 6-day program
                    done sporadically for the vast majority of non-competitive athletes."
  },

  "progressive_overload": {
    "definition": "Systematically increasing training stimulus over time to force adaptation",
    "methods":    ["Add weight to the bar",
                   "Add reps at same weight",
                   "Add sets",
                   "Reduce rest period",
                   "Improve movement quality"],
    "tracking":   """
      def log_workout(exercise, sets, reps, weight):
          session = {
              "date": today,
              "exercise": exercise,
              "volume": sets * reps * weight,
              "top_set": max_weight_lifted
          }
          compare_to_last_session(session)
          if no_progress_in_3_weeks:
              flag_for_program_adjustment()
    """
  },

  "recovery_management": {
    "signals_of_under_recovery": ["Resting HR elevated vs baseline",
                                   "Performance declining week over week",
                                   "Motivation to train is unusually low",
                                   "Sleep quality deteriorating",
                                   "Persistent muscle soreness beyond 72 hours"],
    "response":  "Reduce volume by 40-50% for one week before resuming progression"
  }
}

Nutrition for Fitness Goals

NUTRITION_FRAMEWORK = {
  "fat_loss": {
    "principle":  "Caloric deficit is required. Protein is non-negotiable.",
    "target":     "0.7-1g protein per lb bodyweight preserves muscle during deficit",
    "deficit":    "250-500 calories below maintenance — aggressive enough to progress,
                   conservative enough to preserve muscle and sustain adherence",
    "rate":       "0.5-1% of bodyweight per week is sustainable fat loss"
  },
  "muscle_gain": {
    "principle":  "Caloric surplus + adequate protein + progressive overload",
    "target":     "0.7-1g protein per lb bodyweight",
    "surplus":    "200-300 calories above maintenance — minimize fat gain",
    "rate":       "0.25-0.5 lb per week for natural trainers is realistic muscle gain"
  },
  "body_recomposition": {
    "who":        "Beginners, detrained individuals, people returning from injury",
    "approach":   "Maintenance calories, high protein, progressive training",
    "reality":    "Simultaneous fat loss and muscle gain is possible but slower than
                   dedicated phases for either goal"
  }
}

When Life Disrupts the Plan

DISRUPTION_PROTOCOLS = {
  "travel": {
    "hotel_gym":     "Minimum: 30 min, compound movements, maintain frequency",
    "no_gym":        "Bodyweight protocol — push, pull, hinge, squat, carry variations",
    "principle":     "Maintenance beats nothing. One session per week prevents detraining."
  },
  "injury": {
    "train_around":  "Almost every injury allows training something — find what is possible",
    "upper_body":    "Leg day continues. Lower body injury is not a rest day.",
    "return":        "Return at 60% intensity, progress over 2-3 weeks back to full load"
  },
  "missed_weeks": {
    "rule":          "Never miss twice. One disruption is life. Two is a pattern.",
    "return":        "Reduce weight by 20-30%, rebuild over 1-2 weeks — prevents injury"
  }
}

Progress Tracking

METRICS_THAT_MATTER = {
  "performance":    "Weight lifted, reps completed, pace, distance — objective and motivating",
  "body":           "Weekly average weight, monthly measurements, photos every 4 weeks",
  "habits":         "Training sessions completed vs planned — consistency is the metric",
  "energy":         "Subjective energy and mood — leading indicator of program sustainability",
  "avoid":          "Daily scale weight as primary metric — noise overwhelms signal"
}

Quality Check

  • Goal is specific and time-bound
  • Program is matched to actual available time and equipment
  • Progressive overload built into the plan
  • Nutrition targets provided for stated goal
  • Disruption protocols ready for travel and injury
  • Progress tracking is objective and scheduled
Usage Guidance
This skill appears internally consistent and low-risk because it is instruction-only and asks for no credentials or installs. Before enabling, consider privacy: the skill will likely ask for personal health and activity data (weight, injuries, sleep, etc.), so avoid sharing highly sensitive medical records unless you trust the environment. Also remember this is guidance, not personalized medical advice—consult a qualified healthcare professional for medical conditions or injuries.
Capability Analysis
Type: OpenClaw Skill Name: fit Version: 1.0.0 The skill bundle is a structured fitness coaching framework designed to guide an AI agent in providing workout and nutrition advice. It contains no executable code, network activity, or data access requests, and its instructions in skill.md are entirely consistent with its stated purpose as a fitness operating system.
Capability Assessment
Purpose & Capability
The name and description promise a personal fitness operating system; the SKILL.md contains only fitness assessment, program design, nutrition, disruption handling, and tracking guidance. There are no unrelated dependencies, credentials, or install steps requested that would be out of scope for a fitness guidance skill.
Instruction Scope
The runtime instructions are prose and pseudocode describing how to assess users, design programs, and track progress. They do not instruct the agent to read files, access environment variables, contact external endpoints, or collect unrelated system information. They do imply collecting personal health/training data from the user, which is expected for this type of skill.
Install Mechanism
No install specification or code files are present (instruction-only), so nothing will be downloaded or written to disk. This is the lowest-risk install model and matches the skill's nature.
Credentials
The skill declares no required environment variables, credentials, or config paths. The instructions don't reference secrets or external APIs. The amount and type of data the skill needs (user body measurements, habits, constraints) are proportionate to a fitness coaching system.
Persistence & Privilege
The skill is not marked always:true and does not request persistent system privileges. Agent autonomous invocation is enabled by platform default but is not combined with any other risky behaviors here.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install fit
  3. After installation, invoke the skill by name or use /fit
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release
Metadata
Slug fit
Version 1.0.0
License
All-time Installs 1
Active Installs 1
Total Versions 1
Frequently Asked Questions

What is Fit?

Your personal fitness operating system. Not a workout plan. A complete system that figures out what your body actually needs, builds training around your rea... It is an AI Agent Skill for Claude Code / OpenClaw, with 304 downloads so far.

How do I install Fit?

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

Is Fit free?

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

Which platforms does Fit support?

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

Who created Fit?

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

💬 Comments