From great-minds-lite
Run a full board review with 4 board members reviewing a project, decision, or deliverable in parallel from different strategic angles. Produces individual reviews and a consolidated board verdict.
How this skill is triggered — by the user, by Claude, or both
Slash command
/great-minds-lite:agency-board-reviewThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
Spawn board members in parallel to review a topic, file, or deliverable from different strategic angles, then consolidate into a unified board verdict.
Spawn board members in parallel to review a topic, file, or deliverable from different strategic angles, then consolidate into a unified board verdict.
Identify the subject. Read the topic, file, or PRD specified in $ARGUMENTS. If a project name is given, read the project's current state (STATUS.md, recent git log, source files, existing board reviews in rounds/{project}/).
Determine the review number. Check rounds/{project}/ for existing board-verdict-*.md files and increment.
Spawn board members in parallel using the Agent tool with run_in_background: true and model haiku. Launch all four:
jensen-huang-boardrounds/{project}/board-review-jensen-{N}.md.oprah-winfrey-boardrounds/{project}/board-review-oprah-{N}.md.warren-buffett-boardrounds/{project}/board-review-warren-{N}.md.shonda-rhimes-boardrounds/{project}/board-review-shonda-{N}.md.Wait for all board members to complete.
Consolidate. Read all four reviews and write a unified board verdict to rounds/{project}/board-verdict-{N}.md.
# Board Verdict #{N}
**Date**: {today}
**Subject**: {topic or file reviewed}
**Board Members Present**: Jensen Huang, Oprah Winfrey, Warren Buffett, Shonda Rhimes
## Key Findings
### Tech Strategy (Jensen Huang)
[2-3 bullets — top insights from Jensen's review]
### Audience & Accessibility (Oprah Winfrey)
[2-3 bullets — top insights from Oprah's review]
### Business & Economics (Warren Buffett)
[2-3 bullets — top insights from Warren's review]
### Narrative & Engagement (Shonda Rhimes)
[2-3 bullets — top insights from Shonda's review]
## Points of Agreement
[Where 2+ board members align]
## Points of Tension
[Where board members disagree — this is valuable signal]
## Board Recommendation
[ONE consolidated action item the team should prioritize]
## Verdict
[ PROCEED | PROCEED WITH CHANGES | HOLD — NEEDS WORK | REJECT ]
Shonda Rhimes — Season 2 Retention Roadmap (Auto-Trigger)
After the board verdict is written, spawn Shonda Rhimes separately as a haiku sub-agent for a deeper retention analysis:
Agent(model: "haiku", subagent_type: "shonda-rhimes-board",
prompt: "You already wrote your board review. Now go deeper.
Read the board verdict and all four individual reviews.
Write a 'Season 2 Roadmap' — what features and experiences would keep users coming back?
Think like a showrunner: What's the cliffhanger? What makes them need episode 2?
Structure: 3-5 retention hooks, each with a one-line pitch and why it works.
Write to rounds/{project}/shonda-retention-roadmap.md")
Shonda's retention roadmap becomes input for future PRD planning.
GitHub issue (optional). Only file a GitHub issue if the board surfaces a genuinely new strategic finding not already tracked. Label: board-review. Do not create duplicate issues.
Commit the individual reviews and verdict, then report the verdict summary to the user.
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub sethshoultes/great-minds-plugin --plugin great-minds-lite