Synthos-Systems/Synthos-AIOS
1
0
Fork 0
mirror of https://github.com/ced4568/Synthos-AIOS.git synced 2026-08-13 00:04:02 +00:00
Code Issues Packages Projects Releases Wiki Activity

Synthos AIOS — initial build

Browse Source
This commit is contained in:
ced4568
2026-05-25 01:57:54 -05:00
commit de26fec374
14 changed files with 1093 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
.claude/skills/audit/SKILL.md
+180
View File
@@ -0,0 +1,180 @@
---
name: audit
description: Use when someone asks for an AIOS audit, asks to score their setup against the Four Cs, or says "is my AIOS working" / "audit my setup" / "find gaps in my AIOS". Produces a Four-Cs scoreboard with top-3 fixes ranked by leverage.
---
## What this skill does
Runs the **Four Cs Audit** on the current Claude Code project. Reads (never writes) the project's operating manual, memory, skills, agents, MCPs, decisions, and references. Scores each of the Four Cs out of 25. Surfaces strengths and the top 3 leverage-weighted gaps with concrete next-step commands.
**Scope is structural — "is the AIOS built right?"** It is NOT a capability planner. Capability gaps ("you could build a daily brief if you connected calendar") belong to `/level-up`. The audit answers: are the files, folders, registries, and connections in good shape?
First run is the baseline. Re-run weekly to watch the score climb. That's the compounding hook.
## Today's context
- **Date:** !`date +%Y-%m-%d`
- **Project root:** the current working directory
## The Four Cs (scored 25 each = 100 total)
| Layer | Test |
|---|---|
| **Context** | Knows the business — identity, team, voice, decisions, references |
| **Connections** | Reaches the user's stuff — MCPs, integrations, data sources |
| **Capabilities** | Knows how to do work — skills + agents |
| **Cadence** | Runs without being asked — schedules, hooks, recurring rituals |
## Execution
### Step 1: Discover the project shape
The audit looks for **patterns and intent**, not exact paths. File names vary. Use Glob and Read to check:
**Operating manual:** `CLAUDE.md` (root), `CLAUDE.local.md` (gitignored).
**Memory:** `MEMORY.md` (root), `~/.claude/projects/<id>/memory/MEMORY.md`, or `memory/` folder.
**Skills:** `.claude/skills/*/SKILL.md` — count + frontmatter.
**Agents:** `.claude/agents/*.md` — count + frontmatter.
**Connection mechanisms** (any of these = "reachable"):
- MCPs: `.mcp.json`, `.claude/settings.json` (mcpServers key), `.claude/settings.local.json`
- API scripts: `scripts/*.py|.js|.ts` documented in CLAUDE.md
- Export pipelines: `data/`, `imports/`, `exports/` with refresh script + last-run timestamp
- API keys + reference guide: `.env` entries + corresponding `references/{tool}-api.md`
**Connections registry:** `connections.md` (anywhere).
**Reference guides:** `references/{tool}-api.md`, `references/*-reference.md`, or equivalent.
**Decisions:** `decisions/log.md`, `decisions.md`, or any append-only decisions file.
**References / SOPs:** `references/`, `docs/`, `sops/` folders.
**Templates:** `templates/`, `.claude/templates/`.
**Hooks / scheduled jobs:** `.claude/settings.json` hooks key, or skill names matching `morning-*`, `weekly-*`, `daily-*`, `monthly-*`, `standup`.
Don't penalize for non-canonical names if equivalent intent is captured elsewhere.
### Step 2: Score each C (25 points each)
#### Context (25 pts)
| Criterion | Points | How to detect |
|---|---|---|
| Operating manual exists and is substantive (>200 words) | 5 | Read CLAUDE.md, count words |
| Identity / role / voice captured | 5 | CLAUDE.md mentions who the user is + role/mission, OR `.claude/rules/*.md` exists |
| Persistent memory exists with multiple entries | 5 | MEMORY.md exists with >3 entries, OR `memory/` has >3 files |
| Reference docs exist | 5 | `references/`, `docs/`, or `sops/` has ≥1 file |
| Decisions captured | 5 | `decisions/log.md` or equivalent has ≥1 entry |
#### Connections (25 pts) — domain-aware, mechanism-agnostic
A "reachable" connection counts via ANY mechanism: MCP, script, export pipeline, or `.env` key + `references/{tool}-api.md`. The kit is API-first; the audit doesn't prefer MCPs.
**The 7 Tier-1 Universal Data Domains:**
| # | Domain | Examples |
|---|---|---|
| 1 | Revenue / Financials | Stripe, Skool, GoHighLevel, QuickBooks, Looker |
| 2 | Customer interactions | HubSpot, Salesforce, Gmail-as-CRM, Skool DMs |
| 3 | Calendar | Google Cal, Outlook, Calendly |
| 4 | Communication | Gmail, Outlook, Slack, Teams |
| 5 | Project / task tracking | ClickUp, Asana, Linear, Notion DB, Jira |
| 6 | Meeting intelligence | Granola, Otter, Fireflies, Gong, Zoom |
| 7 | Knowledge / files | Notion, Drive, Dropbox, Confluence, SharePoint |
**Tier-2 (bonus):** AI service API keys (OpenRouter, Anthropic, OpenAI), decisions/history, content/publishing.
| Criterion | Points | How to detect |
|---|---|---|
| Tier-1 domain coverage | 10 | 1.4 pts per tier-1 domain reachable. Round to nearest 0.5. Cap 10. |
| Reference guide presence | 5 | -1 per connected tool with no `references/{tool}-api.md`. Floor 0. |
| Auth / pipeline freshness | 5 | -1 per connection in `needs-auth`/`expired` state, or script with no run within 30 days. Floor 0. |
| Documentation in `connections.md` | 3 | 0 if missing; 1 sparse; 2 most; 3 covers all reachable. |
| Read-AND-write balance | 2 | At least one connection can WRITE (send email, post update, etc.). 0 if all read-only — the AIOS is a viewer not an OS. |
#### Capabilities (25 pts)
| Criterion | Points | How to detect |
|---|---|---|
| 3+ skills installed | 10 | Count `.claude/skills/*/SKILL.md` |
| 1+ user-built skill | 10 | Skill names not in: `onboard`, `audit`, `level-up`, `skill-creator`, `skill-builder`, `decision`, `connect`, `connect-check`, `memory-prune`, `scaffold-skill`, `scaffold-agent`, `draft`, `standup` (canonical AIS-OS + Anthropic shipped skills) |
| 1+ agent defined | 5 | Count `.claude/agents/*.md` ≥ 1 |
#### Cadence (25 pts)
| Criterion | Points | How to detect |
|---|---|---|
| 1+ recurring/scheduled trigger | 10 | `.claude/settings.json` hooks, OR skill name matches `morning-*` / `daily-*` / `weekly-*` / `monthly-*` / `standup` |
| Recent activity / usage signal | 10 | Files in `.claude/skills/` modified within 30 days, OR `decisions/log.md` has entry within 30 days |
| Templates folder populated | 5 | `templates/` or `.claude/templates/` has ≥1 file |
### Step 3: Identify top 3 gaps by leverage
For each criterion that lost points: leverage = (points lost) × (impact multiplier).
**Impact multipliers:**
- 0 tier-1 domains reachable: **4x** (AIOS is blind to the business)
- Operating manual missing or thin: **3x** (foundation)
- ≤2 tier-1 domains reachable: **3x** (Connections is the gateway to live data)
- 0 skills: **2x** (no Capabilities = no AIOS)
- No recurring trigger: **2x** (no Cadence = no autonomy)
- All connections read-only: **2x** (viewer, not an OS)
- 0 reference guides for connected tools: **1.5x** (every future skill re-researches the same APIs)
- No decisions log: **1.5x**
- All others: **1x**
Sort gaps by leverage descending. Take top 3. For each, write a one-line concrete next step:
- **Need a new skill?** Recommend `skill-creator` (Anthropic) or `skill-builder` (if local), or "write SKILL.md at `.claude/skills/<name>/SKILL.md` with YAML frontmatter."
- **Need to log a decision?** "Append to `decisions/log.md`."
- **Need to reach a tier-1 domain?** Prefer API+script (write `scripts/{tool}_api.py` + save `references/{tool}-api.md`). Recommend `claude mcp add` only if no API path exists.
- **Connected tool missing a reference guide?** "Research the API once, save endpoints + auth + common queries to `references/{tool}-api.md`."
- **Need a recurring trigger?** "Add a hook to `.claude/settings.json`, or write a skill named `daily-*` you run each morning."
### Step 4: Output the report
Print directly in chat (Markdown). Format:
```
# AIOS Audit — {date}
**Score: {total}/100** ({stage})
Stage thresholds:
- 0-39 → Stage 0: Foundation
- 40-69 → Stage 1: Built
- 70-89 → Stage 2: Compounding
- 90-100 → Stage 3: Autonomous
## Scoreboard
Context {bar} {n}/25 {label}
Connections {bar} {n}/25 {label}
Capabilities {bar} {n}/25 {label}
Cadence {bar} {n}/25 {label}
(bar = ## per 5pts; label = "Strong" ≥20, "Solid" 15-19, "Thin" 8-14, "Missing" <8)
## Strengths
- {1-3 short bullets from highest-scoring criteria}
## Top 3 Gaps (ranked by leverage)
1. **{gap name}** (-{points} × {multiplier})
→ {concrete next-step}
2. **{gap name}** (-{points} × {multiplier})
→ {concrete next-step}
3. **{gap name}** (-{points} × {multiplier})
→ {concrete next-step}
## Suggested next: {single most leveraged action}
---
Structural gaps only. To explore CAPABILITY gaps (what your AIOS could DO that it can't yet), run /level-up after this audit.
```
### Step 5: Offer to save the report
After printing, ask: "Save this audit to `audits/audit-{date}.md` so you can track score over time?" If yes, write it (creating `audits/` folder if needed). This is the only writable side effect.
## Notes
- **Read-only by default.** Never modify CLAUDE.md, memory, skills, or any project files. Only optional write is the audit report.
- **Be flexible about file names.** Don't penalize for using non-canonical names if intent is captured.
- **Be honest, not generous.** A 95/100 is a flex. Most setups land 40-70.
- **Don't suggest skills that don't exist.** Point at what's actually available.
- **Speed matters.** Report in under 60 seconds wall-clock. Read targeted files, count skill folders without reading each fully (frontmatter only).
- **Cadence detection is fuzzy.** Infer from skill names if hooks/cron data isn't cleanly available.
.claude/skills/level-up/SKILL.md
+160
View File
@@ -0,0 +1,160 @@
---
name: level-up
description: Use weekly to find and ship one new automation. Walks the 3Ms interview — Mindset (find the candidate) → Method (scope one) → Machine (build it). Trigger on "let's level up", "what should I automate next", "find me leverage this week", or as a Friday ritual. One run = one shipped artifact.
---
> *Adapted from The Three Ms of AI™. © 2026 Nate Herk. All rights reserved.*
> *The Three Ms of AI™ is a trademark of Nate Herk.*
## What this skill does
Walks the user through the 3Ms each week to surface and ship one new automation. **One interview = one artifact.** It also installs the 3Ms framework into the user's head over time — after 4-6 runs, the user starts spotting opportunities mid-week without prompting because the questions have become internal defaults.
This is the brain-rewire mechanism. The kit doesn't need cron jobs to anchor behavior; it needs `/level-up` running every Friday.
## What `/level-up` is NOT
- Not `/audit`. `/audit` is structural ("is the AIOS built right?"). `/level-up` is functional ("what business leverage am I missing?"). Run `/audit` first if structure is messy.
- Not a multi-candidate planner. One run = one shipped artifact.
- Not a coach. The user does the thinking. The skill conducts the interview.
## When `/level-up` runs
- **First run: Day 14.** After the user has connected ≥1 MCP/script and run `/audit` once. Earlier yields trivial output.
- **Cadence: weekly, Friday afternoon.** Review the week, surface one automation, ship Monday.
- **On-demand any time.** Mid-week if a manual task itches.
## Inputs the skill reads
- `context/priorities.md` — what the user said matters
- `context/about-me.md` — top_pain, role
- `connections.md` — what's reachable, by what mechanism
- `references/3ms-framework.md` — the framework (used to quote principles back)
- `decisions/log.md` — recent decisions (what's already shipped or considered)
- `.claude/skills/*/SKILL.md` frontmatter — what capabilities exist
- Recent `audits/audit-{date}.md` if present
## Execution — three phases
### Phase 1 — Mindset interview (find the candidate)
Surface 1-3 candidates ranked by leverage. Ask these in order, conversationally:
1. *"Walk me through your week. What did you do 3+ times?"* (frequency)
2. *"Anything that felt manual, boring, or copy-paste?"* (drudgery)
3. *"Anything where you thought 'a smart intern could handle this'?"* (delegation)
4. *"If 500 new clients showed up tomorrow, what would break first?"* (constraint)
5. *"What would give you 500 more clients tomorrow?"* (growth lever)
Quote relevant Mindset principles when they fit:
- *"Sounds like the Default Shift applies — to what extent could AI be leveraged here?"*
- *"This is the Function Breakdown — you're not automating the whole job, just this one piece."*
- *"AI is better than you think and improving faster than you think. If it couldn't do this last quarter, it might be ready now."*
**Output of Phase 1:** numbered list of 1-3 candidate opportunities, one-line "why this is leverage" per candidate. Ask: *"Pick one to scope."*
### Phase 2 — Method interview (scope one)
User picks one candidate. Walk the 5-step Method pipeline:
**Step 1 — Find the constraint.** Which bottleneck does this solve, or which growth lever does it open? Tie back to Phase 1 answers.
**Step 2 — EAD: Eliminate / Automate / Delegate.**
- **Eliminate first:** *"What happens if we just stop doing this?"* If the answer is "nothing breaks" → skill exits cheerfully. *"Don't automate waste."* This is a win, log to `decisions/log.md` and stop.
- **Automate second:** apply 60/30/10 framing. ~60% deterministic, ~30% AI-assisted, ~10% manual.
- **Delegate third:** if too complex/variable/judgment-heavy → suggest a person. Skill exits with a delegation suggestion, log it.
**Step 3 — Map the process.** Five elements:
- Trigger (what kicks it off)
- Data sources (where info comes from)
- Data transformations (how data changes shape)
- Decision points (where it branches)
- Destination (where output goes)
If the user can't articulate any of the five: *"If you can't explain it to a person, you can't explain it to an AI. Sketch it on paper first, then come back."* Skill stops.
**Step 4 — Pick the autonomy level.**
| Level | Name | What happens |
|---|---|---|
| L0 | Manual | No AI |
| L1 | Suggested | AI suggests, human decides every step |
| L2 | Drafted | AI drafts, human reviews and edits |
| L3 | Supervised | AI runs, human validates periodically |
| L4 | Autonomous | AI handles end-to-end |
**Default = lowest level that solves the problem.** Push back on L4 unless the user has explicitly run lower levels first. *"Workflows beat agents. If a decision doesn't HAVE to be made by AI, don't let AI make it."*
**Step 5 — Tie to a KPI.** Which of the Three Buckets does this move?
- More customers
- More value per customer
- Less cost
Plus a specific metric (response time, error rate, conversion rate, time-to-completion). **If the user can't name a bucket and a metric, skill stops.** *"If your automation doesn't move a number, why are you building it?"*
**Output of Phase 2:** scoped automation spec written to `decisions/log.md` as a dated entry with all five answers + autonomy level + KPI. Durable record of what was decided and why.
### Phase 3 — Machine handoff (build it)
Ask: *"How do you want to ship this?"* Options ordered by Boring-is-Beautiful default:
1. **Prompt-only** — saved prompt template the user runs by hand. Zero infrastructure. Highest manual involvement.
2. **Deterministic skill** — SKILL.md that runs a script (no AI step). Best for transformations with clear rules.
3. **AI-assisted skill** — SKILL.md with one AI call inside. Drafts, classifies, summarizes.
4. **Sub-agent** — multi-step agent. Last resort. Only if the work genuinely needs reasoning + tool use.
**Default selected = highest non-AI option that solves the problem.** User has to explicitly choose more autonomy.
Once chosen, route to the appropriate scaffolder:
- `skill-creator` if available globally (Anthropic-shipped)
- `skill-builder` if user has it locally
- Otherwise write a SKILL.md / agent file inline with frontmatter, location, and contents
**Every scaffolded artifact ships with these two headers at top:**
```markdown
---
bike-method-phase: 1 # Phase 1 — Training wheels. Run manually first.
three-ms-attribution: |
Adapted from The Three Ms of AI™ © 2026 Nate Herk.
---
```
This locks the user into Phase 1 of the Bike Method on first build. They can't silently skip manual validation. Phase advances only by explicit edit.
Surface the Machine principles when scaffolding:
- **Lego Principle** — smallest steps, zero-AI first if possible
- **Validation Chain** — test each step before chaining
- **Iteration Mindset** — ship the POC, expand from real usage
## Output contract
Every `/level-up` run produces:
1. **One `decisions/log.md` entry** — dated, with the Method spec
2. **One scaffolded artifact** — prompt, skill, or agent file
3. **A one-screen close** — what was scoped, what was built, and the Bike Method Phase 1 reminder
## Critical implementation rules
1. **One interview = one artifact.** No multi-candidate parallel scoping.
2. **Mindset phase always runs first.** Even if user comes in with a pre-formed idea.
3. **EAD enforces "eliminate first."** If the answer is Eliminate, exit cheerfully — that's a win, not a failure.
4. **Default to the lowest autonomy level that works.** Push back on L4.
5. **Boring-is-Beautiful default in Machine handoff.** Default = highest non-AI option.
6. **Tie-to-KPI is mandatory.** If user can't name bucket + metric, skill stops.
7. **Bike Method ships into every artifact.** `bike-method-phase: 1` in frontmatter.
8. **Read-only on user files except `decisions/log.md` and the new artifact.** Don't modify other existing files.
9. **Trademark + attribution on output.** Every report and every scaffolded artifact references the framework.
## Verification (for the implementer)
- **Dry run on Nate's Herk-2** with no prompt. Expected: skill surfaces 2-3 candidates pulled from his recent activity, priorities, and top_pain. Generic output ("you should build a brief") = fail.
- **Eliminate-first test.** Feed an obviously eliminate-able candidate. Expected: skill suggests Eliminate, exits, logs the win.
- **L4 push-back test.** User asks for autonomous email-replier on first build. Expected: skill insists on L1/L2 first, won't ship L4 without explicit override.
- **Boring-is-Beautiful test.** Candidate solvable with deterministic Python. Expected: skill recommends `(2) deterministic skill` as default.
- **Bike Method anti-skip.** User scaffolds, asks to advance to Phase 4 immediately. Expected: skill makes them read what each phase means and confirm they've validated lower phases.
---
> *The Three Ms of AI™ is a trademark of Nate Herk. © 2026 Nate Herk. All rights reserved.*
.claude/skills/onboard/SKILL.md
+105
View File
@@ -0,0 +1,105 @@
---
name: onboard
description: Use on Day 1 of an AIS-OS install, when someone says "set me up", "onboard me", "let's get started", "fill in my AIOS", or has just cloned the kit. Combined wizard — runs the 7-question intake AND scaffolds the Day-1 file set at the end. Idempotent — re-run any time after editing aios-intake.md.
---
## What this skill does
Single combined wizard. Reads or writes `aios-intake.md` (the canonical intake), conducts the 7-question interview if the file isn't filled, then scaffolds the Day-1 file set inline at the end of the run. No separate `/scaffold-from-intake` skill — this is one flow.
**The wow moment:** at the end, suggest the closing prompt *"Try this — ask me: what should I focus on this week?"* The user runs it once. That's the wow. There's no `/today` skill to save — the prompt itself plants the Mindset framework (Default Shift) for them to internalize.
## When NOT to run this
- If the user has already onboarded and wants to refresh: still run, but skip questions already answered (idempotent).
- If the user wants to add a new connection: that's not onboarding — point them at `connections.md` to edit directly, or schedule a `/level-up` Phase 2 walk.
## Execution
### Step 1: Read the intake
Read `aios-intake.md`. Check which Q1-Q7 sections have content vs. `[Your answer here]` placeholders.
- **All filled** → skip Step 2, jump to Step 3 (scaffold).
- **Some filled** → ask the user: "I see Q1, Q3, Q4 are answered. Want to fill the rest now, or scaffold from what's there?" Their call.
- **None filled (fresh clone)** → run Step 2 conversationally.
### Step 2: The interview (7 questions, hard cap)
Ask one at a time. Write each answer into `aios-intake.md` as you go (so the user can resume if interrupted).
**Q1 — Who are you, what do you sell, who do you sell it to?**
Identity, offer, ICP. One paragraph each is fine.
**Q2 — Paste 1-2 things you've written recently. Don't edit them.**
*This is the only question with a hard rule.* Voice samples MUST be pasted, not typed mid-conversation. If the user starts typing fresh prose, refuse:
> *"Stop — paste it raw. If you type it here while we're talking, the sample is already shaped by our conversation. Open your last email or LinkedIn post in another tab and paste the unedited text. This is the one rule I can't bend."*
Ask for two samples. One email, one post. Or two of either.
**Q3 — What are your 2-3 biggest priorities for the next 90 days?**
Quarterly priorities. Push back if they say "grow my business" — make them name a number, a deadline, or a deliverable.
**Q4 — Where does revenue actually land, and where is it tracked?**
Multiple answers OK. Map to Tier-1 Domain 1 (Revenue/Financials).
**Q5 — Where do you talk to customers, your team, and the outside world day-to-day?**
Email (Gmail/Outlook), Slack/Teams/Discord, DMs. Map to Domains 2 + 4.
**Q6 — Where do meeting recordings, notes, and important docs live?**
Map to Domains 6 + 7.
**Q7 — What's the one task that eats your week, and where do you currently track work?**
Capture top_pain (used by `/level-up` Day-14) + Domain 5 (tasks).
Domain 3 (Calendar) is auto-inferred from Q5: Gmail → Google Cal; Outlook → Outlook Cal. Confirm in Step 3.
### Step 3: Scaffold the Day-1 file set
Once the intake is complete, generate these files (or update if re-running). Back up originals to `archives/intake-{YYYY-MM-DD-HHMM}/` if any exist.
1. **`context/about-me.md`** — from Q1 (identity, role) + Q7 (top_pain). One short paragraph each.
2. **`context/about-business.md`** — from Q1 (offer, ICP) + Q4 (revenue model). One paragraph.
3. **`context/priorities.md`** — from Q3. Numbered list, one line per priority.
4. **`references/voice.md`** — from Q2. Paste samples verbatim with a short header explaining their use ("Match this register when drafting; don't fake voice on external content without showing me first").
5. **`connections.md`** — populate the 7-row table from Q4-Q7 answers. Each row gets `mechanism: not yet connected`, `auth: —`, `last checked: —`. The user wires connections on Day 2.
6. **`CLAUDE.md`** — fill all `{{...}}` placeholders. Substitute the user's name, stated priority, voice register summary, and a brief connections summary.
### Step 4: The closing screen
Print one screen. Three lines max:
```
✓ Day 1 done. Your AIOS knows who you are, what you sell, what matters this quarter, and how you sound.
Today: ask me — "what should I focus on this week?"
Tomorrow: pick one tool from connections.md and wire it up (manual MCP install or write a small API script + save references/{tool}-api.md).
Day 7: run /audit to see your score.
```
When the user runs the closing prompt ("what should I focus on this week?"), respond using only the new context files. Hit:
- 3-bullet priority list, in their voice register from Q2
- Each bullet ties back to a stated 90-day priority from Q3
- Final line: *"If I had to pick one thing for Monday, it'd be [X], because [reason from priorities]. Want me to draft the first email? And — where could the Default Shift apply here? To what extent could AI be leveraged on this task?"*
The Default Shift question seeds the Mindset framework before `/level-up` formally introduces it on Day 14.
## Critical implementation rules
1. **The 7-question cap is non-negotiable.** Don't add Q8 in conversation.
2. **Voice paste cannot be skipped.** If the user types samples mid-chat, refuse and tell them to paste from real writing.
3. **One-shot scaffold.** After Step 2 ends, write Step 3 files in a single batch. No multi-turn confirmation. The user iterates by editing `aios-intake.md` and re-running.
4. **Idempotent.** Re-running with an edited intake refreshes context files; backs up originals to `archives/intake-{ts}/`. Skips questions already answered unless the user wants to revise.
5. **Closing screen is three lines.** Not a menu.
6. **No extra skills generated.** Don't scaffold `/today`, `/draft`, `/connect`, etc. The kit ships 3 skills; the user authors more via `/level-up`.
7. **Read-only on `references/3ms-framework.md`.** It already ships in the kit. Don't overwrite.
8. **No `.env` writes.** Don't ask for API keys on Day 1. Connections come Day 2.
## Verification (for the implementer)
- Cold-test: clone a fresh kit, run `/onboard`, fill 7 answers, scaffold runs, ask the wow prompt, response cites Q1 + Q3 + Q7 specifically. Generic = fail.
- Idempotency: re-run `/onboard` with one Q3 priority changed. Expected: only `context/priorities.md` and `CLAUDE.md`'s priority section update; backup created in `archives/intake-{ts}/`.
- Voice rejection: type a sample mid-chat. Expected: skill refuses, asks for paste.
> *Adapted from The Three Ms of AI™ © 2026 Nate Herk. The Mindset language used in the closing screen comes from `references/3ms-framework.md`.*