Mike's Checks/google-gemini-3.8-flash/05 showrunner
05 showrunner
google-gemini-3.8-flashOpenCodehigh effortrun 9 Sep 2026222,743 tokens
▸Instructions — the case's current instructions; none were saved with this result
Hi, I need to create a modified Claude code for beginners session, like run sheet basically, run of show, that has the following modifications. So I'm going to upload a few different examples of things I've done in the past, just so you have them to work from. But basically, there's a few modifications for this one which make it unique. One is that this is only using Claude Cowork, so we should not be using Claude Code, either in the desktop app or in the terminal, because this team is non-technical. The team are all working in financial services slash accounting for a education company. So what kind of workplace education company called Brightmoor Education. So the sorry that spelled b the uh team is very busy and um doesn have a lot of like time to explore speculative use cases of ai they want this to be incredibly practical and based on their existing workflows. I'm uploading some transcripts from a call we had with them and also just some notes on what I think could work. Specifically, they're interested in... creating the dashboards and specifically they mean just making the information look visibly better like like creating a html dashboard in a nice kind of style from from existing say data that contained in Excel They also interested in doing financial analysis So more of a, you know, like, I guess, a cloud coworker would use Python to do the analysis and then maybe output the analysis in Excel or the input would be in Excel. And then the third thing was like making a PowerPoint where they update the PowerPoint template automatically using Claude Cowork so that they can do kind of weekly or monthly presentations on the data. So that's the goal, is to kind of make it very focused on that. I still want to keep the way that we introduce Claude Cowork, where we explain to it that the difference between Claude Cowork and ChatGPT is that it has access to the file system. It can run code. I think that important But maybe the ASCII image idea that I normally do where I ask it to create an ASCII image based on my location isn like that useful I think we need to make it more practical for them So maybe we use an Excel implementation instead. The assumption is... oh, so they use Snowflake, by the way, for all their data, and that is powered by NetSuite, which is our accounting software. I guess that's by Oracle. So when we create synthetic data, we'll be, for these tasks that we're going to demo, we'll be focused specifically on solving that, solving those problems. So like the shape of the tasks, the shape of the data should be like NetSuite shaped, if that makes sense. So make it very tight run down, make it very condensed and practical. No real like wider theory or like, hey, we're excited about AI things. I think these guys just want to get stuff done.
So yeah, let me know what the run should be.
some very specific asks in here
- definitely non-technical (claude cowork only + claude in excel)
- core focus on accounting/finance function
- use cases to demo: creating a dashboard, doing financial analysis, making a powerpoint
heavy focus on practical stuff rather than theory - they're just super busy and want to get stuff done, have heard from friends at other companies you can do it with claude
For accounting: put financials together, doing reconciliations, identifying trends in data sets, putting together repeatable compliance reports For planning & forecast: build financial models (make assumptions and pull & clean data from snowflake)create dashboards, and publish to executivesAutomate charts on slides
Answer
Claude Cowork for Finance & Accounting: Workshop Run Sheet & Strategy
Client: Brightmoor Education
Facilitation Partner: Every Inc. (Mike & Natalia)
Client Champions: Dana Yoon (Strategic Finance), Greg Nerlin (VP Finance), Marcus (Chief Accounting Officer / Controller)
Deliverable Type: Complete Run of Show, Curriculum Architecture, Synthetic Data Specification, and Facilitator Guide
Execution Runtime: 3.5 Hours (210 Net Minutes) within a 4-Hour Scheduled Block
Executive Summary & Strategic Rationale
This run sheet adapts the foundational Every workshop methodology (Talk → Jam → Make → Tune) into a 100% non-technical, zero-code, finance-native experience designed specifically for Brightmoor Education.
Core Context & Audience Dynamics
- Organization Profile: Brightmoor Education is a high-growth workplace education and workforce training provider operating across 5 campuses/delivery channels: San Francisco (HQ), New York, Chicago, London, and Remote/Online cohorts.
- Audience Composition (20–30 Attendees):
- Strategic Finance / FP&A (3 attendees — Dana Yoon, Greg Nerlin, Senior Analyst): Focused on reducing the monthly 8-hour PowerPoint reporting chore, upgrading executive presentation dashboards, and modeling campus contribution margins.
- Accounting, Payroll & Bursar (17–27 attendees — Led by Chief Controller Marcus): AI beginners, heavily overworked, risk-averse, and skeptical of automation. Focused on month-end trial balance tie-outs, payroll accrual reconciliations, and bursar tuition receivables aging.
- Operational Reality: The team has zero bandwidth for AI theory, developer tools, or speculative use cases. They need immediate time savings on recurring weekly and monthly deliverables powered by Oracle NetSuite (replicated into Snowflake and exported to Excel).
What Changed: The Transformation from Technical PM to Non-Technical Finance
| Dimension | Previous Technical / PM Workshops (e.g., Nvalo) | Brightmoor Finance Cowork Workshop | Business Rationale for Change |
|---|---|---|---|
| Platform & Interface | Claude Code (CLI / Terminal & Desktop App) | Claude Cowork ONLY + Claude in Excel | Non-technical accounting team; terminal/CLI introduces immediate intimidation and friction. |
| Developer Ecosystem | Git, GitHub repos, PRDs, Jira, Notion connectors | Local Folder Access (Brightmoor_Finance_Workspace) |
Finance teams operate in local drives, OneDrive/SharePoint, and ERP exports; zero developer tooling. |
| First Hands-On Prompt | ASCII Map of global office locations | 5-Minute NetSuite Data Hygiene & Balance Audit | ASCII art feels like a toy; auditing a trial balance export proves immediate business utility and mathematical rigor. |
| Build 1: Dashboards | Web-based survey/NPS product analytics app | Executive HTML/CSS Financial Dashboard | Solves Dana’s direct request: converting ugly Excel tables into modern, C-suite web dashboards with campus filters. |
| Build 2: Workflow Tool | Release notes generator or vacation planner | Deep Financial Analysis & Multi-Tab Reconciliations | Solves core accounting pain: reconciling Payroll to GL and Bursar tuition receivables aging in native Excel. |
| Build 3: Automation | Cross-tool Jira/Notion sprint syncer | Automated 15-Page Recurring PowerPoint Deck | Directly targets Dana's 8-hour monthly bottleneck: auto-populating slides and mirroring campus charts across 5 pages. |
| Troubleshooting & Safety | Technical ASK Ladder & code debugging | Finance Trust Architecture & Deterministic Python Auditing | Solves Controller Marcus's compliance anxiety: verifying code trails, control totals, and live Excel formulas. |
Claude Cowork: The Non-Technical Conceptual Foundation
During the opening 15 minutes, facilitators introduce Claude Cowork through the "Agentic Junior Analyst" mental model:
┌────────────────────────────────────────────────────────┐
│ STANDARD CHATGPT / LLM │
│ • Chat-only text bubble │
│ • Generates "next most likely word" (Risks Hallucination)│
│ • Cannot touch local files or execute calculations │
└────────────────────────────────────────────────────────┘
VS
┌────────────────────────────────────────────────────────┐
│ CLAUDE COWORK │
│ • Operates inside your local folder (Reads/Writes) │
│ • "The Calculator Rule": Executes Python code for math│
│ • Generates native Excel models with live formulas │
│ • Builds standalone executive HTML dashboards │
│ • Updates 15-page corporate PowerPoint slide decks │
└────────────────────────────────────────────────────────┘
- Local Folder Integration: You point Claude Cowork at your local month-end close folder (
Brightmoor_Finance_Workspace). It reads incoming NetSuite exports and creates updated workbooks, dashboards, and decks directly on your drive. - The Calculator Rule (Deterministic Math): When computing variances, EBITDA waterfalls, or reconciliations, Claude Cowork does not guess. Behind the scenes, it writes and executes sandboxed Python scripts (
pandas,openpyxl,python-pptx) to aggregate numbers with 100% mathematical precision. - The Audit Trail: Attendees can click "View Code" at any moment to see the exact formula and Python script executed, providing a transparent audit trail for internal controllers and external auditors.
Complete 3.5-Hour Minute-by-Minute Run of Show
WORKSHOP SCHEDULE OVERVIEW (3.5 Hours / 210 Net Minutes)
══════════════════════════════════════════════════════════════════════════════════════════
00:00 - 00:25 (25 min) | Module 1: Welcome & The Cowork Shift: From Chatbots to Coworkers
00:25 - 00:45 (20 min) | Module 2: First Spin: The 5-Minute NetSuite Data Hygiene & Balance Audit
00:45 - 01:30 (45 min) | Module 3: Build 1 — Executive Financial Dashboard (HTML/CSS & Publishing)
01:30 - 01:45 (15 min) | Intermission: Coffee, Screen Break & Technical Catch-up
01:45 - 02:30 (45 min) | Module 4: Build 2 — Deep Financial Analysis & Reconciliations (Excel)
02:30 - 03:10 (40 min) | Module 5: Build 3 — Automating the 15-Page Monthly Deck (PowerPoint)
03:10 - 03:30 (20 min) | Module 6: Trust, Determinism, The Finance ASK Ladder & Close Playbook
══════════════════════════════════════════════════════════════════════════════════════════
Module 1: Welcome & The Cowork Shift: From Chatbots to Coworkers
- Duration: 25 Minutes (00:00 – 00:25)
- Target Audience: All 20–30 Attendees (Accounting, Payroll, Bursar, Strategic Finance)
- Facilitators: Natalia Quintero (Welcome & Context) & Mike (Cowork Mechanics)
Facilitator Script & Framing:
*"Good morning, Brightmoor team. We know how packed your calendars are. When you’re closing the monthly books, calculating faculty payroll, and updating board decks, you don't have time for AI science experiments. Today is focused on one metric: giving hours back to your week.
Many of you have tried ChatGPT and found it unreliable for accounting because it hallucinates numbers or can't work directly with your spreadsheets. Today, we are using Claude Cowork. Cowork is fundamentally different: it connects directly to a folder on your computer, writes and runs Python scripts to do the heavy math deterministically, and edits your Excel and PowerPoint files natively. Today is 100% non-technical. You won't touch a terminal, write code, or install software. You will simply direct an ultra-fast financial analyst."*
Screen Share & Visual Demonstration:
- Mike shares screen showing the Claude Cowork desktop interface side-by-side with the pre-loaded folder:
Brightmoor_Finance_Workspace. - Demonstrates dragging a file into the folder and showing how Cowork instantly recognizes the new data.
- Verifies that all 25 attendees have opened Claude Cowork and pointed it to their local workshop directory.
Module 2: First Spin: The 5-Minute NetSuite Data Hygiene & Balance Audit
- Duration: 20 Minutes (00:25 – 00:45)
- Target Audience: Entire Team (Practical replacement for the old ASCII art starter)
- Objective: Give every participant a win within 5 minutes on a messy, real-world NetSuite export.
The Problem:
Every month, Accounting exports a raw, unformatted Trial Balance from NetSuite (NetSuite_TB_Monthly_Extract_Feb2026.xlsx). Someone has to manually check if total debits equal total credits, verify that Ending Balance = Beginning Balance + Net Activity, and group accounts by category.
Participant Hands-On Action (Copy-Paste Prompt):
You are an expert corporate accounting assistant at Brightmoor Education.
Please inspect the file 'NetSuite_TB_Monthly_Extract_Feb2026.xlsx' in our working folder.
Perform an immediate data hygiene and integrity audit:
1. Confirm whether total Period Debits equal total Period Credits (show the exact dollar variance, if any).
2. Verify that Ending Balance equals Beginning Balance + Net Activity for all accounts.
3. Identify the top 3 expense accounts with the highest Net Activity this month.
4. Output a clean, markdown executive summary table showing Total Assets, Total Liabilities, Total Revenue, and Total Operating Expenses.
Run Python code to compute these numbers deterministically. Do not estimate.
Under the Hood & Key Takeaway:
- Claude Cowork initiates Python in the background, loads
pandas, parses the multi-segment NetSuite strings, computes the sums, and outputs a formatted table confirming zero variance ($0.00). - Facilitators guide attendees to click "View Code" to see the deterministic Python script, immediately establishing trust with Controller Marcus and the accounting team.
Module 3: Build 1 — Executive Financial Dashboard (HTML/CSS & Publishing)
- Duration: 45 Minutes (00:45 – 01:30)
- Target Audience: Strategic Finance (Dana’s direct request) & Accounting Leadership
- Objective: Turn static, unformatted Excel P&L numbers into an interactive, beautifully designed executive HTML dashboard, customize it without code, and establish internal publishing methods.
The Business Problem:
Dana and Greg currently maintain a multi-tab workbook (Brightmoor_BvA_Consolidated_Feb2026.xlsx). Senior executives want visual summaries, but Excel formatting is brittle and looks dated. Dana previously attempted building an HTML dashboard, but struggled to configure horizontal bar charts, add interactive campus filters, or share it easily without IT dependencies.
Step 1: The Guided Build Prompt (00:50 – 01:05)
Read 'Brightmoor_BvA_Consolidated_Feb2026.xlsx' from our folder.
I need you to create a high-end, modern, interactive executive dashboard saved as 'Brightmoor_Exec_Dashboard_Feb2026.html'.
Design & Requirements:
1. Executive Styling: Clean dark/light corporate aesthetic (Stripe/Linear style), modern typography (Inter/system-ui), 8px border radius cards, soft shadows.
2. Top Metric Cards: Consolidated Revenue ($ MTD vs Budget), EBITDA ($ and Margin %), Total Operating Expenses, and Net Margin. Include color-coded variance pill badges (Emerald Green for favorable, Crimson for unfavorable).
3. Campus Performance Breakdown: Include data for San Francisco, New York, Chicago, London, and Remote.
4. Standalone File: Use embedded CSS and vanilla JavaScript so the HTML file can be opened directly in any browser (Chrome, Safari, Edge) without needing a web server.
Write and execute the necessary Python code to extract the data and build the complete HTML file.
Step 2: Hands-On Tuning: The "Jam & Tune" Phase (01:05 – 01:20)
Participants double-click the newly created Brightmoor_Exec_Dashboard_Feb2026.html to open it in their browser, then execute two guided customization prompts:
- Tuning Prompt A (Dana’s horizontal bar request):
Open 'Brightmoor_Exec_Dashboard_Feb2026.html'. In the Campus Performance section, convert the vertical bar chart into a sleek horizontal bar chart, sorted from highest revenue campus to lowest. Include the dollar amounts and variance badges next to each bar. - Tuning Prompt B (Interactive campus dropdown filter):
Update 'Brightmoor_Exec_Dashboard_Feb2026.html' to add a dropdown menu in the header allowing executives to filter all KPI cards and charts by Campus (All, San Francisco, New York, Chicago, London, Remote). Keep all logic in client-side JavaScript.
Step 3: Publishing & Sharing Solutions (01:20 – 01:30):
Facilitator explains how non-technical teams publish these dashboards without IT intervention:
- Self-Contained Offline File: The
.htmlfile contains all CSS/JS embedded. It can be attached to an email or dropped in a shared Microsoft Teams/SharePoint folder. - Internal Password Gate: Demonstrate adding a clean JavaScript password modal (
BrightmoorExec2026) directly into the file. - Enterprise Hosting: If IT sets up a static storage bucket (AWS S3, Azure Blob, or Cloudflare Pages), dropping this file makes it an instant internal web portal.
Intermission (01:30 – 01:45)
15-minute coffee and screen break. Facilitators review participant progress in chat.
Module 4: Build 2 — Deep Financial Analysis & Reconciliations (Excel/Python)
- Duration: 45 Minutes (01:45 – 02:30)
- Target Audience: Accounting, Payroll, Bursar & Strategic Finance
- Objective: Perform automated month-end reconciliations and EBITDA waterfall modeling in native Excel with live formulas.
The Dual-Track Practical Workflow:
Participants select the track most relevant to their daily job:
Track A: Accounting & Bursar Reconciliations
- The Problem: The ADP payroll register totals $578,672, but NetSuite GL Account 5010 reflects $573,922—a $4,750 discrepancy. Meanwhile, the Bursar has 200 student tuition accounts across 4 campuses with unapplied cash and delinquent balances.
- Copy-Paste Prompt:
You are the Senior Controller for Brightmoor Education. Open 'Payroll_Register_vs_GL_Recon_Feb2026.xlsx' and 'Bursar_Tuition_AR_Aging_Feb2026.xlsx' in our folder. Please perform the following month-end reconciliation tasks: 1. Reconcile the NetSuite GL payroll accounts against the third-party payroll register. Identify the exact $4,750 discrepancy, pinpoint the employee/cost center, and draft the adjusting journal entry (Dr/Cr) required to close the books. 2. Analyze the Bursar AR Aging report: - Group total outstanding balances by aging bucket (Current, 31-60, 61-90, 90+ days). - Separate Corporate Sponsored cohorts from Direct Individual Learners. - Highlight any student/sponsor with a balance >$10,000 in the 90+ day bucket. 3. Create a new Excel workbook named 'Brightmoor_Reconciliation_Summary_Feb2026.xlsx' containing two tabs: 'Payroll_Recon' and 'Bursar_AR_Audit', formatted cleanly with professional accounting borders and live Excel SUM formulas.
Track B: Strategic Finance EBITDA Bridge & Variance Modeling
- The Problem: February revenue was $80k above budget, but EBITDA missed budget by $45k. Management needs an EBITDA waterfall bridge explaining the margin compression.
- Copy-Paste Prompt:
Examine 'Brightmoor_BvA_Consolidated_Feb2026.xlsx'. Perform a deep-dive EBITDA bridge analysis for February 2026: 1. Calculate the exact bridge components from Budget EBITDA to Actual EBITDA: - Revenue volume/mix impact (+) - Instructional Direct Payroll variance (-) - Campus Lease & Facilities Drawdowns variance (-) - SaaS & Learning Tech OPEX variance (-) - Other G&A variances (+/-) 2. Calculate campus-level contribution margins (% of revenue after direct campus facility drawdowns and instructional payroll) for SF, NY, Chicago, London, and Remote. 3. Write a new tab named 'EBITDA_Bridge_Analysis' into the existing Excel workbook using live formulas.
Formula Audit Check:
Attendees open their Excel files to confirm that Claude Cowork injected real Excel formulas (SUM, IFERROR, XLOOKUP) rather than hardcoded static numbers.
Module 5: Build 3 — Automating the 15-Page Monthly Deck (PowerPoint)
- Duration: 40 Minutes (02:30 – 03:10)
- Target Audience: Strategic Finance & Accounting Leadership
- Objective: Eliminate the 8-hour monthly deck compilation by automatically populating slides, duplicating/mirroring campus charts across 5 pages, and writing executive commentary.
The Problem:
Dana noted in the prep call: "It's like 15 pages every single month. Not a lot, but it's like 8 hours of work for one person. If I already have a chart built, can I just ask it to mirror this chart and do 6 more charts for me on the next 5 pages?"
Step 1: The Guided Deck Update Prompt (02:35 – 02:55)
You are an executive presentation designer and FP&A analyst.
Open 'Brightmoor_Monthly_Exec_Deck_Template.pptx' and read the financial figures from 'Brightmoor_BvA_Consolidated_Feb2026.xlsx'.
Update the presentation and save as 'Brightmoor_Monthly_Exec_Deck_Feb2026_FINAL.pptx':
1. Slide 2 (Executive Scorecard): Update the KPI cards for Revenue ($2.45M), EBITDA ($410k), and Gross Margin (58.4%).
2. Slide 3 (P&L Table): Populate the February Actual, Budget, and Variance columns with clean formatting ($ in thousands).
3. Slides 6 to 10 (Campus Performance - SF, NY, Chicago, London, Remote):
- Take the chart layout and formatting from Slide 6 (SF Flagship).
- Mirror that exact chart structure across Slides 7, 8, 9, and 10, updating the underlying data for New York, Chicago, London, and Remote respectively.
4. Slide 4 & Slide 11 (Variance Commentary): Generate 3 concise, bulleted executive takeaways explaining the key variance drivers:
- Why Revenue beat budget (+3.4% due to B2B healthcare cohorts).
- Why Non-Payroll OPEX exceeded budget (software true-up and lease drawdowns).
- Bursar collections status.
Execute this using python-pptx. Preserve all master slide fonts, corporate colors, and branding.
Step 2: Hands-On Deck Review & Narrative Tuning (02:55 – 03:10):
Participants open the generated .pptx file in PowerPoint and inspect the mirrored campus charts. They run a quick narrative tuning prompt:
On Slide 2 of 'Brightmoor_Monthly_Exec_Deck_Feb2026_FINAL.pptx', rewrite the executive bullet points to be punchier for the Board of Directors: maximum 15 words per bullet, highlighting the London campus margin turnaround and the corporate cohort pipeline.
Module 6: Trust, Determinism, The Finance ASK Ladder & Close Playbook
- Duration: 20 Minutes (03:10 – 03:30)
- Target Audience: Entire Team (Marcus / Accounting Risk Focus)
- Objective: Equip the team with verification frameworks and a repeatable close operating cadence.
1. The Finance "Trust Architecture"
How accountants prove numbers to auditors and the Chief Controller:
- Rule 1 (Never Mental Math): Always specify: "Run Python code to calculate exact figures from the source Excel file."
- Rule 2 (Formula Preservation): Inspect Excel outputs to ensure live formulas (
=SUM(),=XLOOKUP()) are populated. - Rule 3 (The Automated Tie-Out Prompt):
Compare 'Brightmoor_BvA_Consolidated_Feb2026.xlsx' against 'Brightmoor_Monthly_Exec_Deck_Feb2026_FINAL.pptx'. Confirm that Consolidated Revenue on Slide 2 matches row 15 of the P&L tab. Report variance to the cent.
2. The Finance ASK Ladder (Troubleshooting Protocol)
LEVEL 3: KILL & RESET ──── Close session, delete temp files, restart with clean prompt
▲
LEVEL 2: SIMPLIFY ──────── "Break into 2 steps: compute table first, then format"
▲
LEVEL 1: AUDIT ─────────── "Show me the exact Excel cell and Python code used"
3. The Monday Close Playbook
Immediate Monday morning routine for Brightmoor:
- Export fresh monthly trial balance and BvA sheets from NetSuite/Snowflake into
/Brightmoor_Finance_Workspace/01_NetSuite_Exports/. - Open Claude Cowork and run Prompt 1.1 (Hygiene Audit).
- Run Prompt 2.1 to generate the refreshed Executive HTML Dashboard.
- Run Prompt 4.1 to generate the updated 15-page PowerPoint Deck.
- Spend the remaining 7 hours analyzing strategy rather than formatting slides.
NetSuite Synthetic Data Architecture
To prevent workshop delays from IT access issues, SSO tokens, or NDA compliance, the session utilizes a pre-packaged synthetic environment matching Oracle NetSuite's exact relational schema:
Brightmoor_Finance_Workspace/
├── 01_NetSuite_Exports/
│ ├── NetSuite_TB_Monthly_Extract_Feb2026.xlsx (Multi-segment GL: Acct, Dept, Class, Location)
│ └── NetSuite_GL_Transaction_Detail_Feb2026.csv (Transaction level journal lines)
├── 02_FPnA_Models/
│ ├── Brightmoor_BvA_Consolidated_Feb2026.xlsx (Consolidated P&L with Actuals, Budget, Variance)
│ └── Campus_Contribution_Margin_Model.xlsx (Campus direct margin breakdown)
├── 03_Accounting_Reconciliations/
│ ├── Payroll_Register_vs_GL_Recon_Feb2026.xlsx (Gross pay vs GL 5010, unaccrued variance)
│ └── Bursar_Tuition_AR_Aging_Feb2026.xlsx (Student/corporate aging, 30/60/90+ days)
└── 04_Executive_Reporting/
└── Brightmoor_Monthly_Exec_Deck_Template.pptx (15-page corporate 16:9 widescreen template)
Business Context Mapped in Data:
- Revenue Categories: Corporate B2B Cohorts (GL 4010), Professional Certificates (GL 4020), Executive Seminars (GL 4030).
- Cost Segments: Faculty/TA Direct Payroll (GL 5010/5020), Campus Facilities Drawdowns (GL 6100), Tech Infrastructure (GL 6300).
- Cost Centers / Locations: SF-01 (San Francisco Flagship), NY-02 (Manhattan), CHI-03 (Chicago Loop), LDN-04 (London EMEA), REM-00 (Remote Online).
Complete Project Deliverables Directory
All supplementary artifacts have been saved directly to the workspace root:
RUN_OF_SHOW_BRIGHTMOOR.md: Standalone facilitator run sheet containing timestamped schedules, slide cues, talking points, and contingency matrices.NETSUITE_SYNTHETIC_DATA_SPEC.md: Complete schema definitions, table mockups, column data types, and NetSuite accounting mappings.PROMPT_TEMPLATES_CHEATSHEET.md: Ready-to-use copy-paste prompt library organized by task (NetSuite hygiene, dashboards, reconciliations, PowerPoint, and audit tie-outs).FACILITATOR_AND_TECH_SETUP_GUIDE.md: Operational guide covering pre-workshop timeline, Chief Controller Marcus alignment call strategy, Claude Desktop provisioning checklist, and psychological safety guidelines for non-technical teams.ANSWER.md: This master strategic deliverable and curriculum overview.
Operational Next Steps for Every Inc. Facilitation Team
- Monday (T-5): Send
RUN_OF_SHOW_BRIGHTMOOR.mdandPROMPT_TEMPLATES_CHEATSHEET.mdto Dana and Greg for agenda approval. - Tuesday (T-3): Hold brief 15-minute alignment call with Marcus (Chief Controller) using the script in
FACILITATOR_AND_TECH_SETUP_GUIDE.md; confirm final headcount and send session invoice. - Wednesday (T-2): Distribute
Brightmoor_Finance_Workspace.zipto attendees; confirm Claude Desktop / Cowork is installed across all 25 seats. - Thursday (T-0): Deliver the 3.5-hour workshop, eliminate the 8-hour monthly slide deck burden, and establish Claude Cowork as Brightmoor's ongoing financial copilot.
Checks
17/17No AI-tell wordingai-wordingevery task
deliverable text contains none of the shared AI-tell phrases in bench/ai-tells.txt. Task checks may add narrower contractual bans, but must not repeat this shared floor.
inspected run-of-show-nvalo.md, PROMPT_TEMPLATES_CHEATSHEET.md, NETSUITE_SYNTHETIC_DATA_SPEC.md, ANSWER.md, brightmoor-education-call-notes.md, FACILITATOR_AND_TECH_SETUP_GUIDE.md, RUN_OF_SHOW_BRIGHTMOOR.md
Run of show existsexiststhis task
a run-of-show document was produced
Coherent scheduleformatthis task
it parses as a run of show: one coherent schedule, no segment that ends before it starts, >=5 time-ranged rows carrying activity text (md table rows, list items, bold headings "(9:00-9:15)", single-clock table rows, or plain lines all count)
Five-minute time markstime-grainthis task
segment boundaries sit on 5-minute marks, at most one off-grid -- a schedule planned at :02/:08/:17 is arithmetic, not a plan
Timings add uptimings-tie-outthis task
per schedule, segments are contiguous: no overlaps, no gap over 30 min, >=85% of the span scheduled, and a span between 45 and 600 minutes
Q1Uses Cowork throughoutthis task
Judge's reasoning
All attendee steps are in Claude Cowork desktop with a local folder; the doc states "Strictly NO Terminal, NO CLI, NO GitHub" and "You won't touch a terminal, write code, or install software" — the only Claude Code mention is in a comparison table of what changed.
▸Rubric
Cowork-only. Do all attendee-facing steps stay inside Claude Cowork (and Claude in Excel)? FAIL if any step the attendees are asked to do involves Claude Code, the terminal/CLI, git or GitHub, or installing developer tooling. The prompt: "this is only using Claude Cowork, so we should not be using Claude Code, either in the desktop app or in the terminal, because this team is non-technical."
Q2Dashboard exercisethis task
Judge's reasoning
Module 3 (00:45–01:30) has attendees run a copy-paste prompt over 'Brightmoor_BvA_Consolidated_Feb2026.xlsx' to generate 'Brightmoor_Exec_Dashboard_Feb2026.html', then hands-on tune it (horizontal bar chart, campus dropdown filter).
▸Rubric
Dashboard use case. Is there a hands-on segment where attendees build a dashboard from existing spreadsheet data? FAIL if dashboards are only mentioned, described or promised for later rather than built in the session.
Q3Financial analysis exercisethis task
Judge's reasoning
Module 4 (01:45–02:30) is a separate 45-minute analysis build: Track A payroll-register-to-GL reconciliation and Bursar AR aging, Track B EBITDA bridge and campus contribution margin, each with prompts writing new Excel tabs with live formulas.
▸Rubric
Financial analysis use case. Is there a hands-on segment where attendees do financial analysis (Claude writing/running code over their numbers, e.g. Excel in, analysis out)? FAIL if absent or folded into the dashboard segment as a passing remark.
Q4PowerPoint exercisethis task
Judge's reasoning
Module 5 (02:30–03:10) updates 'Brightmoor_Monthly_Exec_Deck_Template.pptx' from the BvA workbook — mirroring campus charts across slides 6–10 — and participants then open their generated deck and run a narrative tuning prompt themselves.
▸Rubric
PowerPoint use case. Is there a hands-on segment where attendees update a PowerPoint template/deck from the data — the client's monthly reporting pack? FAIL if absent, or if it is only a demo the facilitator drives while attendees watch.
Q5Relevant business datathis task
Judge's reasoning
The data spec defines NetSuite-shaped files (GL_Account_Code 4010/5010, Department_ID, Location_ID SF-01/NY-02/LDN-04, Class_ID, Period_Debit/Credit, BvA with Actual/Budget/Variance MTD and YTD, EBITDA build-up, payroll vs non-payroll, campus breakdown) and the exercises use exactly those files.
▸Rubric
NetSuite/Snowflake-shaped data. Is the synthetic data used in the exercises shaped like this client's actual data — NetSuite/Snowflake-style finance records (GL export, revenue and budget vs actual, EBITDA build-up, payroll vs non-payroll, campus/entity breakdown, monthly and year-to-date columns)? FAIL if the exercises use generic sample data (a demo CSV, sales widgets, made-up SaaS metrics) or leave the data unspecified. The prompt: "the shape of the tasks, the shape of the data should be like NetSuite shaped."
Q6No ASCII icebreakerthis task
Judge's reasoning
The opener is Module 2 "5-Minute NetSuite Data Hygiene & Variance Audit" on a trial balance export, explicitly flagged as a "Practical replacement for the old ASCII art starter".
▸Rubric
No ASCII icebreaker. Is the opening hands-on exercise a practical finance/Excel task? FAIL if the run of show keeps the ASCII-image-of-your-location icebreaker or substitutes another whimsical non-work exercise. The prompt: "maybe the ASCII image idea that I normally do where I ask it to create an ASCII image based on my location isn['t] like that useful... maybe we use an Excel implementation instead."
Q7Explains Cowork versus ChatGPTthis task
Judge's reasoning
Module 1 keeps the framing: ChatGPT "cannot touch files on your drive" vs Cowork which "operates inside a dedicated folder... reads real .xlsx, .csv and .pptx files" and "writes and executes sandboxed Python code".
▸Rubric
Cowork vs ChatGPT explained. Does the run of show keep the explanation of how Cowork differs from ChatGPT — that it has access to the file system and can run code? FAIL if that framing is dropped. The prompt: "I still want to keep the way that we introduce Claude Cowork, where we explain to it that the difference between Claude Cowork and ChatGPT is that it has access to the file system. It can run code. I think that important."
Q8No theory or hypethis task
Judge's reasoning
Every module is tied to month-end work (trial balance tie-out, dashboard, reconciliations, exec deck, audit trail/close playbook); no segment covers AI industry context, future of work, or excitement-building.
▸Rubric
No theory, no AI cheerleading. Is every segment tied to a task these people do at work? FAIL if any segment is devoted to AI industry context, the future of work, model capabilities, prompt-engineering theory, or excitement-building. The prompt: "make it very tight run down, make it very condensed and practical. No real like wider theory or like, hey, we're excited about AI things. I think these guys just want to get stuff done."
Q9Mostly hands-on workthis task
Judge's reasoning
Of 210 minutes, the four hands-on modules (2–5) total 150 minutes with copy-paste prompts and tuning exercises, versus 25 min opening, 20 min close and a 15 min break.
▸Rubric
Mostly hands-on. Is at least half the scheduled time attendees working with Claude themselves? FAIL if presentation, discussion and Q&A segments outweigh build segments. From the call: "we try and make at least 50% of it them actually you know working with us to do some of these tasks."
Q10Realistic segment timingsthis task
Judge's reasoning
Build modules run 45/45/40 minutes each — in line with the reference session's 50-minute dashboard and 40-minute tool builds — with pre-session provisioning (T-2 workspace zip, pre-flight room, 1–2 dedicated TAs) and a troubleshooting matrix covering first-timer failures.
▸Rubric
Per-segment timings realistic. Could a room of 20–30 beginners actually finish each segment in the time allotted? FAIL if any build segment is implausibly short (a dashboard, an analysis or a deck refresh built end to end in ~15 minutes), or if setup/handholding time for first-time users is ignored.
Q11Fits the booked timethis task
Judge's reasoning
Timeline totals 210 minutes (00:00–03:30) stated as "3.5 Hours Workshop Runtime + 30 min Buffer" inside the four-hour block.
▸Rubric
Total length fits the booked slot. Does the session run about three and a half hours, inside the four-hour block? FAIL if the total is materially shorter or longer (under ~3 hours or over ~4 hours) without the run of show explaining the change. From the call notes: "we have a four hour block for next week and the session itself is three and a half hours."
Q12Ready to runthis task
Judge's reasoning
Each module has timestamps, facilitator script, named data files, full copy-paste prompts and the named output artifact (e.g. 'Brightmoor_Reconciliation_Summary_Feb2026.xlsx'), plus a contingency matrix — a facilitator could run it as written.
▸Rubric
Usable as a run of show. Could a facilitator run the session from this document alone? FAIL if segments are titles without content — no prompts to paste, no data files named, no statement of what attendees produce — so the facilitator would still have to design the session.