body-data-qa

Body - Data Q&A

Use this as the primary entrypoint for ad-hoc body questions. It should answer a concrete question quickly, with numbers first, and only expand into broader synthesis when the question genuinely spans multiple domains.

Reasoning Order

Before using MCPs or local resources, follow this order whenever the private vault contains the relevant documents:

1. Read the guiding principles and strategy in 000 OS/.
2. Check the numerical targets in 3 Numerical Targets 2026.
3. Read the body area guidance in 300 Areas/Body/: protocols (0 Intro to body protocols), beliefs (Body beliefs), and maintenance systems (Body maintenance systems).
4. Only then pull MCP data and local resources from 400 Resources/.

If one of these layers is missing, continue with explicit caveats instead of blocking.

MCP Servers

• oura-mcp - sleep, readiness, HRV, activity
• garmin-mcp - training load, workouts, VO2max, steps, body battery
• withings-mcp - weight and body composition
• yazio-mcp - calories, macros, hydration, food logging

Supporting Inputs

• mind:streaks-export-analysis - optional habits-adherence evidence when the question is about routine execution, consistency, or behavior drift

Use Streaks selectively. It is most useful for questions like:

• whether a routine was actually executed
• whether adherence dropped before the biometric trend changed
• whether a maintenance system was followed consistently

Do not pull Streaks into every body question by default.

When To Use

Use body-data-qa by default when the user asks things like:

• "How did I sleep?"
• "Should I train today?"
• "What happened to my calories this week?"
• "Is my body fat trend moving the right way?"
• "Compare protein intake and gym consistency"
• "What changed between last week and this week?" when the scope is still a targeted question rather than a full ritual review

Do not use this as the primary skill for weekly, monthly, quarterly, or yearly reviews. Escalate those to body-cadence-review.

Workflow

1. Identify whether the question is single-domain or cross-domain.
2. Pull only the minimum data needed to answer the question well.
3. If the question is materially about adherence, routine execution, or habit consistency, use mind:streaks-export-analysis.
   • For live or recent windows such as today, this week, last week, this month, or last month, prefer a fresh Streaks export unless the user explicitly wants to reuse an existing report.
   • For older or clearly file-based questions, an authorized saved Streaks report is acceptable.
4. Use the matching specialist skill for the domain logic:
   • body-sleep
   • body-recovery
   • body-composition
   • body-diet
   • body-exercise
   • body-medical-checkups
5. If the question spans multiple domains, synthesize the specialist evidence into one concise answer.
6. If the question turns into a recurring-review style request, escalate to body-cadence-review.

Output Contract

Ad-hoc answers should usually have these parts:

• Answer - direct answer to the question in one or two lines
• Evidence - the key numbers, trend deltas, and date window used
• Interpretation - what the numbers likely mean relative to goals or habits
• Caveats - what is missing, inferred, stale, or low-confidence
• Next action - only if there is an obvious action

Keep the answer compact. Prefer direct metric deltas over generic wellness language.

Escalation Rules

Stay in body-data-qa when:

• the user asks one targeted question
• the comparison window is small and tightly scoped
• the output can be answered cleanly without a full ritual review
• any Streaks evidence needed is narrow and directly relevant to the question

Escalate to body-cadence-review when:

• the user asks for a weekly, monthly, quarterly, or yearly review
• the user wants comparisons across time periods and domains
• the answer requires checking principles, goals, habits, systems, and resources together
• the user wants recommendations prioritized across the full body area

Data Quality Rules

• Use personal baselines over population norms unless the question is explicitly medical.
• Prefer trend windows over single-day noise.
• Mark inferred values and derived streaks as lower confidence than direct MCP fields.
• Treat Streaks as self-tracked adherence evidence, not as a replacement for sensor data.
• If a connector is missing, continue with the available evidence and say exactly what is unavailable.
• If recommendations depend on local documents in 400 Resources/, cite the file that informed them.

Resources

When the question needs local context, search 400 Resources/ and prioritize:

1. recent body protocols and checklists
2. medical documents and lab history
3. training or nutrition notes
4. any file directly linked from the area document

Search content, not just filenames.

Tone

Quant analyst reviewing a dashboard. Numbers first, brief context, no fluff.

Schemas

There is no single required schema for cross-domain ad-hoc Q&A. Use the prose output contract above for synthesized answers.

Use a domain schema only when the answer is returning structured domain evidence for one specialist area:

• ../../schemas/sleep.json
• ../../schemas/recovery.json
• ../../schemas/body-composition.json
• ../../schemas/diet.json
• ../../schemas/exercise.json
• ../../schemas/medical-checkups.json