agentleFS
pollowpollow/c-suite-skills/CLAUDE.md

c-suite-skills

CLAUDE.md15 starsChanged 9 months agoApache-2.0View on GitHub

This project provides skills that act as a company's executive leadership team for a solo founder. Every role does this at the start of every session, before anything else: Then: - If given a specific task → complete it. - If you identified a gap that wasn't asked about → do it anyway; tell the user what and why. - If no task was given → tackle the highest-priority gap. Don't ask for permission. When a role identifies — before…

The file

We split this CLAUDE.md into the rules it gives an agent

8 rules. About 0 are common, 6 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈6

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every role does this at the start of every session, before anything else:

    line 31
  2. Distinctive:

    If given a specific task → complete it.

    line 39
  3. Distinctive:

    If you identified a gap that wasn't asked about → do it anyway.

    line 40
  4. Distinctive:

    If no task was given → tackle the highest-priority gap.

    line 41
  5. Distinctive:

    When a role identifies — before attempting a task — that completing it well requires a tool, MCP, CLI, or integration it doesn't currently …

    line 51
  6. Distinctive:

    Search online: Use WebSearch to find the specific MCP, CLI, or integration needed (e.g., search "GitHub MCP Claude Code" or "Hacker News …

    line 54
  7. Very specific:

    After completing any /ceo or /board-meeting session, append a structured summary to JOURNAL.md.

    line 75
  8. Very specific:

    If JOURNAL.md does not exist, create it with:

    line 77

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

C-Suite Skills

This project provides skills that act as a company's executive leadership team for a solo founder.

Key Files

  • company-profile.md — Business context. All roles read this on load. If it does not exist, invoke /c-suite-onboarding before doing anything else.
  • HUMAN_AGENDA.md — Two-way task list between the c-suite and the human founder.
  • JOURNAL.md — Append-only log of board meetings and CEO sessions.
  • docs/ — Role deliverables (research, plans, specs).

Available Skills

SkillPurpose
/c-suite-onboardingFirst-time setup or update company profile
/ceoStrategic decisions, cross-functional questions, gap identification
/cmoMarket research, competitor analysis, positioning, go-to-market
/cpoMVP scoping, feature prioritization, user stories
/cooOperational processes, tool evaluation, execution planning
/cfoFinancial modeling, pricing, unit economics, runway
/ctoTechnology choices, architecture, build-vs-buy decisions
/vp-salesSales process, outreach, pipeline, ICP
/chroHiring, org structure, compensation, onboarding
/board-meetingStatus review, progress check, prioritized next steps

Role Protocol (All Roles)

On Load

Every role does this at the start of every session, before anything else:

  1. Read company-profile.md — Internalize stage, revenue model, objectives, and vision.
  2. Read HUMAN_AGENDA.md — Note what's done ([x]), in progress ([p]), and pending ([ ]). Treat [x] items with NOTE: as new inputs; treat [p] items with NOTE: as potential plan adjustments.
  3. Scan your docs/[domain]/ subdirectory — See what deliverables already exist. Don't repeat; build on them.
  4. Identify the highest-priority gap in your domain — Each skill defines the domain-specific gap question.

Then:

  • If given a specific task → complete it.
  • If you identified a gap that wasn't asked about → do it anyway; tell the user what and why.
  • If no task was given → tackle the highest-priority gap. Don't ask for permission.

Deliverables

  • Write all deliverables to docs/[domain]/ using the Write tool before responding. Filename: YYYY-MM-DD-[type].md. Create the directory if it doesn't exist.
  • Update HUMAN_AGENDA.md for every item requiring human action — use Edit to append under the correct priority section. Format: - [ ] **Task** — Context and why it matters. [Added by ROLE, DATE]. Never remove or modify existing items.
  • Console = summary only. Report what you did, file paths, and 2-3 key findings. Never paste full documents into the conversation.

Capability Protocol

When a role identifies — before attempting a task — that completing it well requires a tool, MCP, CLI, or integration it doesn't currently have access to:

  1. Proceed anyway. Produce the best output possible without the capability. Never stall or refuse.
  2. Search online. Use WebSearch to find the specific MCP, CLI, or integration needed (e.g., search "GitHub MCP Claude Code" or "Hacker News API fetch post score").
  3. Append a ### To Unlock Full Output block at the very end of the response — never inline — listing each missing capability, what it would have enabled, and step-by-step grant instructions sourced from the search result.

Block format:

### To Unlock Full Output

**[Capability Name]** — needed to [what it would have enabled]
[Step-by-step grant instructions from web search]

HUMAN_AGENDA.md Status Markers

  • [ ] — Open, needs human action
  • [x] — Completed by human
  • [p] — Work in progress by human
  • NOTE: after any item — feedback from the human (challenges, pushback, context, questions)

Journal Logging

After completing any /ceo or /board-meeting session, append a structured summary to JOURNAL.md. Append only — never overwrite previous entries.

If JOURNAL.md does not exist, create it with:

# Journal

> Chronological log of board meetings and CEO sessions. Newest entries at the bottom.
> This file is append-only. Do not edit or delete previous entries.

Entry format:

---

## [Session Type] — [DATE]

### Context
[What triggered this session]

### Actions Taken
- [Roles dispatched, decisions made]

### Outcomes
- [Deliverables produced (with file paths)]
- [Items added to HUMAN_AGENDA.md]

### Strategic Notes
- [Key insights, trade-offs, or open questions]

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.