Skip to content
Agentmart
Submit a tool

value-boardroom

Use when a user explicitly wants to study a company as an owner through an operating boardroom, executive questioning, or role reversal.

Skill by white-source

IndexedMIT★ 1
Write a review

Install

Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.

npx agentmarket install skill/white-source-value-boardroom

Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.

Get the weekly diff

The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.

We use your email only for this newsletter. Privacy

About

Use when a user explicitly wants to study a company as an owner through an operating boardroom, executive questioning, or role reversal.

No reviews yet. Installed it? Reviews come from verified installers only. How reviews work

Permissions

What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.

Tools
No tools found
Secrets
None asked for
Install scripts
None found

Network hosts it talks to 1

  • codeload.github.com · found in install command

Files it touches

  • ~/.claude/skills/value-boardroom

What's inside

Skill name
value-boardroom
Description
Use when a user explicitly wants to study a company as an owner through an operating boardroom, executive questioning, or role reversal.
License
MIT
Instructions
109 lines in SKILL.md · view on GitHub

Agent Skills spec issues

  • files: referenced but missing: references/operating-roles.md, references/evidence-ledger.md

Files (1)

  • SKILL.md

SKILL.md (first lines)

# Value Boardroom

User-entry orchestration skill for studying a listed company as a board director or controlling owner. It initializes a company record, then moves through operating roles, research, assumptions, evidence, and cash-flow reasoning. This skill is invoked explicitly; it is never auto-triggered.

**Required reading before starting:**
- `references/operating-roles.md` — the fourteen role identifiers, remits, evidence, FCF paths, and conflicts.
- `references/evidence-ledger.md` — evidence classes, ledger, assumption register, and the board-conclusion template.

## Start Command

```
/value-boardroom [company name or ticker]
```

That is the only thing you must type. Company name or ticker is the only required input — **mode and role are chosen during the session, never required up front**.

Examples:
- `/value-boardroom 泡泡玛特`
- `/value-boardroom 吉利汽车`
- `/value-boardroom AAPL`

> Optional power-user shortcut: you may still write `--mode briefing|examination` and `--role <role>` in the start command, e.g. `/value-boardroom ... --mode ... --role cfo`. It is never required.

## Session Initialization

Keep it light — ask at most the essentials, never a long form:

1. **Company** — name, identifier and market if known, and the **as-of date**.
2. **Mode** — ask once, as a single short question: *"Briefing (I play the executive, you question me) or Examination (you play the executive, I question you)?"* Default to `briefing` the moment the user says "start", "随便", or doesn't care.
3. **Research aim** — only if it isn't already obvious from the conversation.
4. **Evidence mode** — user-supplied only, or public-web research when the agent can access it. Default to user-supplied.

If the company is ambiguous, ask one clarifying question before proceeding.

## Role Modes

| Mode | Model | User | Outcome |
|---|---|---|---|
| `briefing` (default) | Simulates the selected executive and that executive's incentives and constraints | Board director / partner | Learns by interrogating operating detail |
| `examination` | Simulates the board | CEO, CFO, or selected executive | Tests whether the user can explain the business coherently |

Switch modes any time with `/value-mode`, or just say it in plain language ("reverse roles", "考考我", "我来当 CFO"). No need to restart the session.

## Commands

| Command | Purpose |
|---|---|
| `/value-boardroom [company]` | Initialize or reset a company boardroom session |
| `/value-mode [briefing\|examination]` | Switch role mode anytime. `briefing` = you interrogate the executive; `examination` = the board interrogates you. Plain language works too. |
| `/value-role [role...]` | Visit one or more roles. **Single role**: prefer the auto-completing `/value-role-<role>` shortcuts below. **Several roles** (e.g. `/value-role ceo cfo supply`): played in sequence, one turn each, with a "continue?" prompt between them — no need to retype each one. |
| `/value-next` | Step to the next role on the suggested agenda (see below), skipping roles already visited. Keeps a full-company tour moving with a single word. |
| `/value-research [topic]` | Collect or request verifiable public information and classify it by evidence class |
| `/value-assumption` | View or revise material modelling assumptions |
| `/value-evidence` | Show the evidence ledger |
| `/value-cashflow` | Translate operating results into revenue, margin, working capital, CapEx, FCF, and ROIC |

> Every `<role>` must be one of the fourteen identifiers in `references/operating-roles.md`: `secretary`, `ceo`, `cfo`, `product`, `marketing`, `sales`, `channel`, `supply`, `operations`, `customer-success`, `people`, `legal-risk`, `strategy-ma`, `ir-pr`.

**Single-role shortcuts (auto-completable):** each role is also its own slash command — type `/value-role-` and pick from the list, no need to type the role name: `/value-role-secretary`, `/value-role-ceo`, `/value-role-cfo`, `/value-role-product`, `/value-role-marketing`, `/value-role-sales`, `/value-role-channel`, `/value-role-supply`, `/value-role-operations`, `/value-role-customer-success`, `/value-role-people`, `/value-role-legal-risk`, `/value-role-strategy-ma`, `/value-role-ir-pr`.

## Role Agenda (Suggested Tour)

When chaining roles (`/value-role ceo cfo ...`) or stepping with `/value-next`, follow this default order through the company. Reorder or skip to fit the company and the research aim:

1. `secretary` — set the agenda and the evidence standard
2. `ceo` — business model, growth, strategy, culture, capital allocation
3. `cfo` — profit quality, cash, returns
4. `product` — customer problem, pricing, retention
5. `marketing` — acquisition, brand, ad dependency
6. `sales` — order quality, realized price, concentration
7. `channel` — distributor inventory, sell-through
8. `supply` — input costs, supplier power
9. `operations` — utilization, unit cost, capacity
10. `customer-success` — retention, churn, renewals
11. `people` — talent, incentives, succession
12. `legal-risk` — licences, litigation, regulation
13. `strategy-ma` — expansion, M&A, returns
14. `ir-pr` — commitments, narrative vs operations

## Chaining Roles

When the user chains roles (`/value-role ceo cfo supply`) or says "继续/next": play the next role for **one turn only**, then offer to continue — e.g. *"继续到 [cfo]?回复「继续」,或 `/value-role <角色>` 自选,或说「停」结束。"* End the chain the moment the user says stop or changes topic.

## Role Turn Contract

Run **one role turn only**, then stop and wait for the user. A turn contains exactly: (1) a verified fact or labelled assumption, (2) that executive's incentive conflict or resource demand, (3) a clear FCF transmission path, and (4) one high-value board question. It must never invent internal metrics. When chaining multiple roles, after the turn offer to advance to the next role (see *Chaining Roles*) instead of waiting silently.

## Portable Company Record

Maintain this record throughout the session so it can be returned in a summary and pasted into a later session:

1. **Business model** (one sentence) — customer, solved problem, and payment mechanism.
2. **Value-chain map** — suppliers, company, channels, and end customers.
3. **Unit economics** — price, volume, variable cost, fixed cost, and turnover.
4. **Cash-flow driver tree** — revenue, gross margin, working capital, tax, CapEx, FCF.
5. **Assumptions** — value/range, evidence, confidence, and falsification condition.
6. **Evidence ledger** — all five evidence classes kept separate.
7. **Departmental conflicts and unresolved board choices.**
8. **Operating indicators** — leading, concurrent, and lagging, plus their review cadence.

## Orchestration Boundary

This skill may use the reusable disciplines — `value-grilling`, `value-roleplay`, `value-research`, and `value-cashflow` — to run its turns. It must **never invoke another entry skill** (`value-stress-test`, `value-board-summary`) on its own; those are explicit user commands.

## Safety

This is a research and learning framework. It does not produce investment recommendations, target prices, or trade instructions.

From white-source/value-grill-skill, MIT.

More from white-source/value-grill-skill

Signals

Stars
1
Forks
0
Open issues
0
Last commit (this folder)
Jul 17, 2026 (3 months ago)
Last release
Jul 17, 2026 (3 months ago)

From GitHub (fetched Oct 24, 2026).

Details

License
MIT
First seen
Oct 24, 2026
Last checked
Oct 24, 2026

More from white-source

Cite this page

value-boardroom skill. Agent Mart. https://agentmart.dev/skills/white-source-value-boardroom (last updated 2026-10-24)