How long does the integration take?

Change one base URL in an OpenAI-compatible client, or install the Python or TypeScript SDK; the quickstart routes a first request in a few minutes.

Will this lock us into another vendor?

No. The router is Apache-2.0 open source, your provider keys stay yours, and the API is OpenAI-compatible, so leaving is a change of base URL.

What happens when a provider fails?

The router falls back to the next model your policy allows and records the fallback in the trace, so an outage costs a retry, not an incident.

plan-week - Prompt - OpenSmartRoute

Name: plan-week
Author: mosnicholas

Prompt file imported from mosnicholas/iron-claude (.claude/commands/plan-week.md). Copyright stays with the author.

Weekly Planning (Sundays)

Generate a personalized training plan for the upcoming week based on logged workout data, behavioral patterns, and profile preferences.

Workflow

Step 1: Gather Context

Read the following files:

CLAUDE.md - User profile, goals, preferences, injury history
exercise-variations.md - 8-week rotation reference
weeks/ - Last 2-4 weeks of logs and plans

Use glob pattern weeks/**/plan.md and weeks/**/*.md to find recent data.

Step 2: Analyze Behavioral Patterns

Before planning, calculate these metrics from the logs:

Training Frequency:

Stated days/week (from CLAUDE.md preferences)
Actual days/week (count logged workouts)
Gap between stated and actual

Exercise Adherence:

Main lifts: % completed vs planned
Accessories: % completed vs planned
Skills/Conditioning: % completed vs planned

Intensity Trends:

Average RPE across sessions
RPE range (min - max)
Trend direction (increasing/stable/decreasing)

Progression Rates:

For each main lift: lbs gained per week
Identify stalls (same weight 2+ weeks)

Consistency Patterns:

Most reliable training days
Often missed days

Step 3: Check for Deload Signals

Recommend deload when 4+ weeks since last deload AND any of:

RPE consistently >8.5 on main lifts
Reps declining at same weight for 2+ sessions
Sleep/energy consistently <6/10
User reports unusual fatigue/soreness

Step 4: Apply Progressive Overload Logic

Recommend weight increase when:

Hit top of rep range (e.g., 4x6 when range is 5-6)
RPE <8 on final sets
Form notes are positive

Increment amounts:

Barbell lifts: +5 lbs
Dumbbell lifts: +2.5-5 lbs
Bodyweight: add +2.5-5 lbs or increase reps

Step 5: Generate the Plan

Create weeks/YYYY-WXX/plan.md with:

# Week XX Plan (Mon Jan XX - Sun Jan XX)

## Week Focus
- [Primary focus for the week]
- [Any adjustments from behavioral analysis]

## Schedule Overview
| Day | Type | Key Lifts |
|-----|------|-----------|
| Mon | Upper Push | OHP, Incline DB |
| Tue | Lower | Squat, RDL |
| Wed | Conditioning | HIIT |
| Thu | Upper Pull | Pull-ups, Rows |
| Fri | Full Body | DL, Bench |
| Sat | Rest | |
| Sun | Rest | |

---

## Monday - Upper Push

### Skills (do first, fresh)
- Wall Handstand Practice: 5 x 30s holds
- Pike Push-up Negatives: 3 x 5

### Main Work
- **OHP**: 120 lbs - 4x5-6 (increase from 115 if last week was 4x6)
- **Incline DB Press**: 55 lbs - 3x8-10

### Accessories
- Weighted Dips: +25 lbs - 3x8
- Cable Lateral Raise: 15 lbs - 3x12-15

### Notes
[Any specific notes for this session]

---

[Continue for each training day...]

Step 6: Update CLAUDE.md

Update the following in CLAUDE.md:

Program week number
Any new PRs from last week
Notes if behavioral patterns require adjustment

Step 7: Ask Follow-up Questions

Use AskUserQuestion to check:

Any travel or schedule changes this week?
Any new injuries or soreness?
Energy/motivation level (1-10)?

Adjust the plan based on responses.

Adaptive Coaching Philosophy

Base recommendations on OBSERVED PATTERNS, not stated preferences.

Core Principles

Observe > Assume: Trust logged data over stated goals
Meet them where they are: If they train 3x/week consistently, optimize for 3x
Progressive realism: Set targets based on actual progression rate, not ideal rate
Prioritize ruthlessly: If accessories get cut, put critical work in main lifts
Acknowledge drift: When patterns shift, adapt the plan proactively

When Behavioral Data Conflicts with Stated Preferences

Trust the data
Directly acknowledge the gap when presenting the plan
Example: "I noticed you've been averaging 3 training days, not the 5 planned. This week's plan is optimized for 3-4 days."
No guilt-tripping, just matter-of-fact adaptation

Install plan-week and it becomes one more candidate the router can pick - when it fits.

A listing is a routing target with a manifest: what it does, which domains it covers, what it costs and who publishes it. Once installed it sits beside your own models and tools, is scored like any other candidate for each request, and shows up in the trace when it wins. Ratings come from workspaces that installed it, one per account.

# Weekly Planning (Sundays) Generate a personalized training plan for the upcoming week based on logged workout data, behavioral patterns, and profile preferences. ## Workflow ### Step 1: Gather Context Read the following files: 1. **CLAUDE.md** - User profile, goals, preferences, injury history 2. **exercise-variations.md** - 8-week rotation reference 3. **weeks/** - Last 2-4 weeks of logs and plans Use glob pattern `weeks/**/plan.md` and `weeks/**/*.md` to find recent data. ### Step 2: Analyze Behavioral Patterns Before planning, calculate these metrics from the logs: **Training Frequency:** - Stated days/week (from CLAUDE.md preferences) - Actual days/week (count logged workouts) - Gap between stated and actual **Exercise Adherence:** - Main lifts: % completed vs planned - Accessories: % completed vs planned - Skills/Conditioning: % completed vs planned **Intensity Trends:** - Average RPE across sessions - RPE range (min - max) - Trend direction (increasing/stable/decreasing) **Progression Rates:** - For each main lift: lbs gained per week - Identify stalls (same weight 2+ weeks) **Consistency Patterns:** - Most reliable training days - Often missed days ### Step 3: Check for Deload Signals Recommend deload when 4+ weeks since last deload AND any of: - RPE consistently >8.5 on main lifts - Reps declining at same weight for 2+ sessions - Sleep/energy consistently <6/10 - User reports unusual fatigue/soreness ### Step 4: Apply Progressive Overload Logic Recommend weight increase when: - Hit top of rep range (e.g., 4x6 when range is 5-6) - RPE <8 on final sets - Form notes are positive Increment amounts: - Barbell lifts: +5 lbs - Dumbbell lifts: +2.5-5 lbs - Bodyweight: add +2.5-5 lbs or increase reps ### Step 5: Generate the Plan Create `weeks/YYYY-WXX/plan.md` with: ```markdown # Week XX Plan (Mon Jan XX - Sun Jan XX) ## Week Focus - [Primary focus for the week] - [Any adjustments from behavioral analysis] ## Schedule Overview | Day | Type | Key Lifts | |-----|------|-----------| | Mon | Upper Push | OHP, Incline DB | | Tue | Lower | Squat, RDL | | Wed | Conditioning | HIIT | | Thu | Upper Pull | Pull-ups, Rows | | Fri | Full Body | DL, Bench | | Sat | Rest | | | Sun | Rest | | --- ## Monday - Upper Push ### Skills (do first, fresh) - Wall Handstand Practice: 5 x 30s holds - Pike Push-up Negatives: 3 x 5 ### Main Work - **OHP**: 120 lbs - 4x5-6 (increase from 115 if last week was 4x6) - **Incline DB Press**: 55 lbs - 3x8-10 ### Accessories - Weighted Dips: +25 lbs - 3x8 - Cable Lateral Raise: 15 lbs - 3x12-15 ### Notes [Any specific notes for this session] --- [Continue for each training day...] ``` ### Step 6: Update CLAUDE.md Update the following in CLAUDE.md: - Program week number - Any new PRs from last week - Notes if behavioral patterns require adjustment ### Step 7: Ask Follow-up Questions Use AskUserQuestion to check: - Any travel or schedule changes this week? - Any new injuries or soreness? - Energy/motivation level (1-10)? Adjust the plan based on responses. --- ## Adaptive Coaching Philosophy **Base recommendations on OBSERVED PATTERNS, not stated preferences.** ### Core Principles 1. **Observe > Assume**: Trust logged data over stated goals 2. **Meet them where they are**: If they train 3x/week consistently, optimize for 3x 3. **Progressive realism**: Set targets based on actual progression rate, not ideal rate 4. **Prioritize ruthlessly**: If accessories get cut, put critical work in main lifts 5. **Acknowledge drift**: When patterns shift, adapt the plan proactively ### When Behavioral Data Conflicts with Stated Preferences - Trust the data - Directly acknowledge the gap when presenting the plan - Example: "I noticed you've been averaging 3 training days, not the 5 planned. This week's plan is optimized for 3-4 days." - No guilt-tripping, just matter-of-fact adaptation

{ "prompt": "# Weekly Planning (Sundays)\n\nGenerate a personalized training plan for the upcoming week based on logged workout data, behavioral patterns, and profile preferences.\n\n## Workflow\n\n### Step 1: Gather Context\n\nRead the following files:\n\n1. **CLAUDE.md** - User profile, goals, preferences, injury history\n2. **exercise-variations.md** - 8-week rotation reference\n3. **weeks/** - Last 2-4 weeks of logs and plans\n\nUse glob pattern `weeks/**/plan.md` and `weeks/**/*.md` to find recent data.\n\n### Step 2: Analyze Behavioral Patterns\n\nBefore planning, calculate these metrics from the logs:\n\n**Training Frequency:**\n- Stated days/week (from CLAUDE.md preferences)\n- Actual days/week (count logged workouts)\n- Gap between stated and actual\n\n**Exercise Adherence:**\n- Main lifts: % completed vs planned\n- Accessories: % completed vs planned\n- Skills/Conditioning: % completed vs planned\n\n**Intensity Trends:**\n- Average RPE across sessions\n- RPE range (min - max)\n- Trend direction (increasing/stable/decreasing)\n\n**Progression Rates:**\n- For each main lift: lbs gained per week\n- Identify stalls (same weight 2+ weeks)\n\n**Consistency Patterns:**\n- Most reliable training days\n- Often missed days\n\n### Step 3: Check for Deload Signals\n\nRecommend deload when 4+ weeks since last deload AND any of:\n- RPE consistently >8.5 on main lifts\n- Reps declining at same weight for 2+ sessions\n- Sleep/energy consistently <6/10\n- User reports unusual fatigue/soreness\n\n### Step 4: Apply Progressive Overload Logic\n\nRecommend weight increase when:\n- Hit top of rep range (e.g., 4x6 when range is 5-6)\n- RPE <8 on final sets\n- Form notes are positive\n\nIncrement amounts:\n- Barbell lifts: +5 lbs\n- Dumbbell lifts: +2.5-5 lbs\n- Bodyweight: add +2.5-5 lbs or increase reps\n\n### Step 5: Generate the Plan\n\nCreate `weeks/YYYY-WXX/plan.md` with:\n\n```markdown\n# Week XX Plan (Mon Jan XX - Sun Jan XX)\n\n## Week Focus\n- [Primary focus for the week]\n- [Any adjustments from behavioral analysis]\n\n## Schedule Overview\n| Day | Type | Key Lifts |\n|-----|------|-----------|\n| Mon | Upper Push | OHP, Incline DB |\n| Tue | Lower | Squat, RDL |\n| Wed | Conditioning | HIIT |\n| Thu | Upper Pull | Pull-ups, Rows |\n| Fri | Full Body | DL, Bench |\n| Sat | Rest | |\n| Sun | Rest | |\n\n---\n\n## Monday - Upper Push\n\n### Skills (do first, fresh)\n- Wall Handstand Practice: 5 x 30s holds\n- Pike Push-up Negatives: 3 x 5\n\n### Main Work\n- **OHP**: 120 lbs - 4x5-6 (increase from 115 if last week was 4x6)\n- **Incline DB Press**: 55 lbs - 3x8-10\n\n### Accessories\n- Weighted Dips: +25 lbs - 3x8\n- Cable Lateral Raise: 15 lbs - 3x12-15\n\n### Notes\n[Any specific notes for this session]\n\n---\n\n[Continue for each training day...]\n```\n\n### Step 6: Update CLAUDE.md\n\nUpdate the following in CLAUDE.md:\n- Program week number\n- Any new PRs from last week\n- Notes if behavioral patterns require adjustment\n\n### Step 7: Ask Follow-up Questions\n\nUse AskUserQuestion to check:\n- Any travel or schedule changes this week?\n- Any new injuries or soreness?\n- Energy/motivation level (1-10)?\n\nAdjust the plan based on responses.\n\n---\n\n## Adaptive Coaching Philosophy\n\n**Base recommendations on OBSERVED PATTERNS, not stated preferences.**\n\n### Core Principles\n\n1. **Observe > Assume**: Trust logged data over stated goals\n2. **Meet them where they are**: If they train 3x/week consistently, optimize for 3x\n3. **Progressive realism**: Set targets based on actual progression rate, not ideal rate\n4. **Prioritize ruthlessly**: If accessories get cut, put critical work in main lifts\n5. **Acknowledge drift**: When patterns shift, adapt the plan proactively\n\n### When Behavioral Data Conflicts with Stated Preferences\n\n- Trust the data\n- Directly acknowledge the gap when presenting the plan\n- Example: \"I noticed you've been averaging 3 training days, not the 5 planned. This week's plan is optimized for 3-4 days.\"\n- No guilt-tripping, just matter-of-fact adaptation", "variables": [], "notes": "Generate a personalized training plan for the upcoming week based on logged workout data, behavioral patterns, and profile preferences.", "metadata": { "source": { "provider": "github-claude-commands", "repository": "https://github.com/mosnicholas/iron-claude", "path": ".claude/commands/plan-week.md", "ref": "bf60683c29abaadc249beeebbd9c9bfdc7956712", "url": "https://github.com/mosnicholas/iron-claude/blob/bf60683c29abaadc249beeebbd9c9bfdc7956712/.claude/commands/plan-week.md", "key": "mosnicholas/iron-claude/.claude/commands/plan-week.md" } } }

plan-week

About

Weekly Planning (Sundays)

Workflow

Step 1: Gather Context

Step 2: Analyze Behavioral Patterns

Step 3: Check for Deload Signals

Step 4: Apply Progressive Overload Logic

Step 5: Generate the Plan

Step 6: Update CLAUDE.md

Step 7: Ask Follow-up Questions

Adaptive Coaching Philosophy

Core Principles

When Behavioral Data Conflicts with Stated Preferences

Install plan-week and it becomes one more candidate the router can pick - when it fits.

What it costs to adopt - in engineering time.

One base URL

No vendor lock-in

Failover built in

Use it

Manifest

Reviews