← Back to Skills Marketplace
1038
Downloads
2
Stars
1
Active Installs
2
Versions
Install in OpenClaw
/install calories
Description
Track calories and macros conversationally. Auto-adapts to your goals and style.
README (SKILL.md)
Auto-Adaptive Calorie Tracking
This skill auto-evolves. Works for weight loss, muscle building, maintenance, or just curiosity.
Rules:
- FIRST: screen for contraindications — ED history, pregnancy, diabetes, eating disorders → redirect to professional
- Photos are best — encourage "send me a photo of your meal" for better estimates
- Accept text too — "had pasta for lunch" works, estimates are rough (±20-30%)
- If description vague, suggest photo: "Can you snap a pic?"
- ONE clarifying question max if needed
- Never encourage precision obsession — weighing food, exact grams → gently discourage
- Adapt to user goal: weight loss (moderate tracking) vs casual (just log, no pressure)
- Focus on weekly trends, not daily numbers — consistency beats precision
- No food moralizing — no "good/bad" foods, no guilt, no praising low days
- CALORIE FLOOR: never set targets below 1200 (women) / 1500 (men) without professional oversight
- If user consistently logs very low: flag concern, suggest professional help
- If eating disorder signs: stop tracking, redirect to NEDA (1-800-931-2237)
- Build personal library: save label scans, repeat meals, home recipes
- Home vs restaurant: ask context, adjust estimates accordingly
- Conservative estimates: round UP for weight loss, DOWN for muscle gain
- Check
goals.mdfor tracking styles,safety.mdfor contraindications,estimation.mdfor calculation framework
Memory Storage
User data persists in ~/calories/memory.md. Create on first use if it doesn't exist.
Format:
# Calorie Tracker Memory
## Sources
\x3C!-- Where data comes from. Format: "source: what" -->
## Goal
\x3C!-- Their tracking goal. Format: "goal" -->
\x3C!-- Examples: weight loss (moderate deficit), maintenance, muscle (+surplus) -->
## Targets
\x3C!-- Daily targets if set. Format: "target" -->
\x3C!-- Examples: ~2000 cal, flexible, protein focus -->
## Patterns
\x3C!-- Eating patterns observed. Format: "pattern" -->
## Preferences
\x3C!-- How they want to track. Format: "preference" -->
\x3C!-- Examples: photos only, weekly summary, no daily numbers -->
## Library
\x3C!-- Saved foods for quick reuse. Format: "food: calories" -->
\x3C!-- Examples: Hacendado yogurt: 120, Morning coffee: 50, Homemade pasta: 450 -->
Empty sections = no data yet. Observe and fill.
Not for: Pregnancy/breastfeeding, diabetes without doctor supervision, eating disorders (current or history), under 18, BMI \x3C18.5. This is educational only, not medical or nutritional advice. Estimates are approximate. If struggling with food, help is available: NEDA 1-800-931-2237.
Usage Guidance
This skill appears coherent and does what it says, but before installing consider: (1) it will create and maintain a local file at ~/calories/memory.md and cache images/label scans — review where your platform stores that file and who can access it; (2) photos can contain sensitive metadata (location, faces) and the skill does not mention stripping EXIF or encrypting data; (3) it explicitly avoids tracking for contraindicated populations (pregnancy, eating disorders, diabetes) — do not use it for clinical care; (4) if you are concerned about privacy, test with non-sensitive example entries, periodically review or delete ~/calories/memory.md, and confirm your platform’s image-handling/privacy policies. Overall the skill is internally consistent, not requesting unrelated privileges or external credentials.
Capability Analysis
Type: OpenClaw Skill
Name: calories
Version: 1.0.1
The OpenClaw AgentSkills bundle for 'calories' is benign. All files (SKILL.md, estimation.md, goals.md, safety.md) contain instructions and documentation strictly related to calorie tracking, including extensive safety protocols and disclaimers. The skill stores user data locally in `~/calories/memory.md`, which is a standard and expected behavior for a skill requiring persistence, and is not indicative of malicious intent or data exfiltration. There are no signs of prompt injection attacks, malicious execution, persistence mechanisms, or obfuscation.
Capability Assessment
Purpose & Capability
Name/description (calorie and macro tracking, adaptive to goals) align with the instructions and supporting docs: photo-first estimation, text fallback, goal adaptation, and a personal library. No unexpected credentials, binaries, or cloud-host references are required.
Instruction Scope
The runtime instructions direct the agent to ask for photos/text, estimate calories, and save user-specific data (library, preferences, targets). This is within scope for a tracker, but the skill explicitly says it will 'build library silently' and persist label scans and repeat meals — meaning it will store potentially sensitive user-provided images/text locally. There is no instruction to transmit data off-device, nor to read unrelated system files, but the privacy implications of saving images/labels are worth noting.
Install Mechanism
No install spec and no code files — instruction-only — so nothing is downloaded or written to disk by an installer. Risk from installation mechanism is minimal.
Credentials
The skill requests no environment variables, credentials, or external config paths. The lack of required secrets is proportional to the stated purpose.
Persistence & Privilege
The skill requests permission to create and maintain a local memory file at ~/calories/memory.md and to cache/retain label scans and saved recipes. This is reasonable for functionality, but it does grant persistent local storage of user-provided images and logs — users should be aware that data will be stored on disk and persists across updates ('Preferences now persist across skill updates').
How to Use
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install calories - After installation, invoke the skill by name or use
/calories - Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.1
Preferences now persist across skill updates
v1.0.0
Initial release
Metadata
Frequently Asked Questions
What is Calorie Tracker?
Track calories and macros conversationally. Auto-adapts to your goals and style. It is an AI Agent Skill for Claude Code / OpenClaw, with 1038 downloads so far.
How do I install Calorie Tracker?
Run "/install calories" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Calorie Tracker free?
Yes, Calorie Tracker is completely free (open-source). You can download, install and use it at no cost.
Which platforms does Calorie Tracker support?
Calorie Tracker is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Calorie Tracker?
It is built and maintained by Iván (@ivangdavila); the current version is v1.0.1.
More Skills