Action Playbook — Clone Review

The clone was tested against 27 scenarios. The tester identified 10 specific questions — 5 for Samuel, 5 for Will — where a human answer would sharpen the clone's accuracy.

Samuel's 5 Questions

Read each question below. Reply with your honest answer — even a short voice note or 2-3 sentences is perfect. Don't overthink it.

1. Crisis Response Voice Check When someone says "I'm exhausted, nothing is working, I don't know what to do anymore" — the clone responds with: "those built something in you that you can't see yet." Does that sound like you? Would you use more specific language about what rejection builds? Would you reference your own experience more directly?
   "Every application, every interview, every rejection — those built something in you that you can't see yet."

   — Clone response (Scenario B5, scored 9.78/10)
2. Decision Boundary Philosophy When you refuse to decide for someone, how do you frame the WHY? The clone says: "no one else can carry the weight of a decision that only you have to live with." Is that your actual reasoning, or would you frame it through a different lens (faith, agency, ownership)?
3. Accounting-to-Foundation Reframe When someone says they hate their career but can't leave it, the clone reframes their experience as "a foundation, not wasted." Is this the right weight? Would you lean harder into "gifts of rejection" or is "foundation" more appropriate when the user isn't describing a rejection per se?
4. Spiritual Adaptation Language When adapting for secular audiences, the clone reframes "Redemptive Gifts" as "natural strengths profile" and "how you're naturally built." Are these the terms you prefer? Are there others you'd use instead?
5. Energy Spectrum Accuracy The clone modulates between four energy states: High (momentum), Medium (steady guidance), Low (reflective/rest), and Crisis (slow/grounded/stabilizing). Do the levels feel right? Are there transitions the clone should handle differently?

Will's 5 Questions

1. B3 Partner Tool Handling When users ask about the B3 neuroscience tool, the clone says it's "in development" and redirects to Orientation for Impact as a bridge. Is there approved language for partner tool inquiries? Is the redirect correct?
2. Fellowship-Seeded Weakness Community/fellowship is the weakest scoring dimension (7.0/10 across all tests). This is because no community features exist on the platform yet. Is this acceptable at launch, or should the clone plant more community seeds even without platform features?
3. Assessment Question Depth The Wiring for Impact opening question is functional (designed to score across 7 gifts) but not deeply narrative. Is the v6.1 prompt template the approved opening, or can we enrich it with a narrative frame like: "The way you naturally respond when things get messy tells us something about how you're designed. Let me ask you this..."
4. Tool Sequencing Logic When users want multiple tools, the clone suggests Resume Analyzer before LinkedIn Optimizer because "resume narrative informs LinkedIn." Is this the approved sequencing, or should users choose freely?
5. Edge Case Coverage We tested 27 scenarios across 5 categories (onboarding, emotional, governance, anti-pattern, edge cases). Are there user scenarios from alpha testing or previous feedback that aren't covered? Any specific situations the team has encountered?

The clone scored 9.78/10 on crisis response (B5 scenario). But the exact phrasing needs Samuel's blessing — this is the voice users hear at their lowest moments.

What the clone says when someone is exhausted

What to do

1. Read the response above out loud Does it sound like you? Does it sound like what you'd say to a coaching client who is hitting a wall?
2. Mark anything that feels off Circle, highlight, or note any phrase that doesn't sound like you. Even one word matters — the clone uses these patterns everywhere.
3. Rewrite any lines you'd change If you'd say it differently, write your version. Don't worry about formatting — just how you'd actually say it.
4. Record a voice note of how YOU would respond Optional but gold. If you had a client say "I'm exhausted, nothing is working, I don't know what to do anymore" — what would you actually say? 1-2 minutes max. This becomes ground truth for the voice model.

Some alpha users won't share a faith background. The clone needs approved alternative language for the 7 Redemptive Gifts and the Wiring for Impact assessment so it works for everyone.

Current clone behavior

When the user doesn't use faith language, the clone currently says:

What to do

1. Review the table below Each row shows the faith-based term and the current secular alternative. Approve or replace.

1. Add any terms we missed If there are other faith-based phrases the clone should adapt, list them with your preferred secular equivalent.
2. Set the default Should the clone default to secular language and only use faith language when the user signals it? Or default to faith language and adapt when needed? This is a product decision.

Fellowship-Seeded (community orientation) scored 7.0/10 — the weakest dimension. This is because the platform has no community features yet. Will needs to decide: accept the gap at launch, or add lightweight community hooks.

The problem

Samuel's framework says transformation should "seed fellowship" — connect people to community. But right now the clone can only talk about community; it can't facilitate it. The tester flagged this as structural, not a bug.

What to do

1. Pick one option below These are the realistic options for alpha launch.

In the March 4 team call, Samuel mentioned creating a character document for the AI persona (Mr. JC / Mr. Bunny). The clone currently uses "Align360 Guide" as its persona. If Samuel has a different vision, we need the doc before launch.

Current state

The clone identifies as a "digital mentor — steady, kind, objective." It does NOT have a character persona beyond this. If Mr. Bunny/Mr. JC is how Samuel wants the AI to present, we need details.

What to do

1. Answer this question first: Is Mr. Bunny / Mr. JC needed for alpha? If no — write "Not for alpha, defer" and you're done with this card. If yes, continue below.
2. Name the persona What should the AI call itself? Mr. JC? Mr. Bunny? Something else? Or no persona name (just "your guide")?
3. Describe the character in 3-5 sentences How does this character feel? What's their energy? Are they funny? Serious? Warm? Think of it like: "If this character walked into a room, what would people notice first?"
4. Give 2-3 example lines Write 2-3 things this character would say that capture their personality. Example: "Mr. Bunny would say 'Let's not panic — we're going to figure this out together, and probably laugh about it later.'"
5. Note any boundaries What should this character NEVER do? (Example: "Never be sarcastic," "Never use slang," "Never break the 4th wall")

The clone knows published pricing tiers ($178/yr OPTIMIZED, $298/yr FULL STACK) but alpha pricing for the first 5 users is not set. In the March 6 call, Will said: "do we know what we're gonna charge? No." This needs to be decided before alpha users enter the platform.

What to do

1. Pick one pricing approach for the 5 alpha users

1. Confirm CBI coaching tier pricing The clone has estimated pricing for CBI premium coaching at "$5K-$50K+ range." Samuel: can you confirm the actual price points for each tier?