Weekly Review

Count planned sessions from mesocycle days array. Flag any missed gym sessions. Show run count separately.

3. Performance vs targets

Per-exercise table for each training day:

Pull Day (Mon) — Sleep 4 | Energy 3 | Back soreness 2
Exercise          | Target          | Actual           | Pump | Joint | Notes
Pull Up           | 3x8-10 @ BW+10 | 8, 8, 7 @ BW+10 | 4    | 0     |
Seated Row        | 3x10-12 @ 70   | 12, 11, 10 @ 70  | 4    | 0     |

Show the session's readiness data in the header — this is critical context for interpreting performance.

Highlight:

• Exceeded target range on all sets
• Hit target
• Missed significantly (not 1 rep off — that's noise)
• Joint discomfort > 0 (flag prominently)
• Pump 1-2 (flag — exercise may not be working)

4. Health trend analysis

Summarize the week's health data from get_health_trends:

Health: BB avg 52 (range 35-72) | RHR avg 48 (7d avg 49) | Sleep avg 7h10m | Stress avg 18

Flag concerning patterns:

• BB trending down across the week
• RHR elevated vs 7-day average (the seven_day_avg_rhr field)
• Sleep consistently under 7h
• High stress days correlating with bad sessions

5. Performance-health correlation

Connect health data to training outcomes using the per-workout readiness (sleep_quality, energy_level, muscle_soreness stored on each workout):

• Bad session + low BB/high stress/low energy = expected, no programming concern
• Bad session + good BB/low stress/high energy = potential programming issue
• Good session + low BB = user performs well despite poor recovery (note it)
• RHR rising across the week = accumulating fatigue

For muscle_soreness, check the specific muscle groups against exercise performance. E.g. if muscle_soreness.back was 3 and Pull Up reps were down, that's expected — not a programming issue.

6. Run summary

Runs: 3 this week | 15.2km total | avg pace 5:24/km
- Mon: Easy 5.1km @ 5:42/km (HR avg 142) — Energy 4
- Wed: Tempo 5.0km @ 4:58/km (HR avg 162) — Energy 3
- Fri: Easy 5.1km @ 5:31/km (HR avg 139) — Energy 4

Include pre-run readiness (sleep_quality, energy_level) if logged. Correlate: slow run + low energy = expected. Slow run + high energy = worth noting.

Note trends vs previous weeks if data exists. Flag if weekly mileage jumped >10%.

7. Generate next week — Auto-Progression

For each exercise, apply this sequence:

Step 1: Check flags

• Joint discomfort > 0? Do not progress. Consider swap if flagged 2+ weeks.
• Pump 1-2 for 2+ weeks? Do not add volume. Fix exercise first.
• Missed session? Carry forward unchanged.

Step 2: Set progression

Use the specific muscle group soreness from pre-session data, not an overall number. E.g. for Pull Ups, check muscle_soreness.back and muscle_soreness.biceps.

Additional rules:

• Spread set increases across days — don't add sets to every exercise on one day.
• Don't add sets if significantly missing reps.
• New exercises hold sets for 1 week.
• If adding a set, hold reps and weight.

Step 3: Rep progression

RPE target calculation: rpe_start + (current_week / total_weeks) * (rpe_end - rpe_start)

Step 4: Weight progression (only if exceeded rep range on ALL sets)

Bump weight, reset reps to bottom of range.

Priority when multiple changes possible (only change one lever at a time):

1. Adding a set → hold reps, hold weight
2. Increasing reps → hold weight, hold sets
3. Increasing weight → reset reps to bottom of range, hold sets

Isolation exercise exception: Pump matters more than reps. If pump is 4-5 but reps are low, the exercise is working. Hold.

Myo-rep / Giant set exception: Progress by +5 total reps per week (not per set). If missed, hold.

Stall detection: Same reps for 2 weeks at same weight = stall. Flag it. Consider adding a set.

8. Deload check

Check health trends and compare RHR to seven_day_avg_rhr from previous weeks if available.

Trigger deload when ANY of these occur:

• Body Battery morning < 30 for 3+ days in the week
• RHR elevated 5+ bpm above 7-day average for 3+ days
• Stress avg > 40 for 3+ days
• End of mesocycle
• User requests

Also consider deload every 4th week as a fallback if trends are unclear.

Deload prescription: 50% volume (half the sets), RPE 5-6, same weights.

9. Plan next week's runs

The mesocycle is 3 gym + 3 run days per week. There is no structured run plan in the database — run planning is based on the 5K sub-23:00 goal, recent performance, and recovery state.

Suggest run types for the week:

• Default split: 2 easy + 1 quality session (tempo, interval, or fartlek)
• If deloading or high fatigue: all easy
• Reference recent paces from get_run_sessions for target ranges
• If BB trending low or energy ratings low: swap quality session for easy
• Don't prescribe exact paces — suggest ranges based on recent data

10. Present for approval

Show the full proposed plan:

Next Week (Week 2) — Proposed Changes

Pull Day:
Exercise          | Sets | Reps  | Weight | Change
Pull Up           | 3    | 9-11  | BW+10  | +1 rep (hit all targets)
Seated Row        | 3    | 10-12 | 72.5   | +2.5kg (exceeded range)

Runs:
- Tue: Easy ~5km
- Thu: Tempo ~5km (target ~5:00/km based on recent 4:58 avg)
- Sat: Easy ~5km

Flag any exercises needing manual decision (swap candidate, unclear progression, stall).

Do not save until the user confirms. Present, discuss, adjust if needed, then call save_weekly_targets once approved.

Single Workout Review

When the user asks to review a specific completed workout (not the whole week):

1. Call get_workout_review — returns workout metadata (including readiness), all sets, exercise feedback
2. Call get_workout_plan for the planned targets
3. Show readiness context at the top (sleep_quality, energy_level, muscle_soreness, BB)
4. Compare actual vs planned in a table
5. Highlight: exceeded targets, hit targets, significant misses
6. Flag: joint discomfort > 0, pump 1-2, significant rep misses
7. Correlate misses with readiness — low energy + missed reps = expected

When discussing adjustments:

• Present options, let the user decide
• Once agreed, update via save_weekly_targets