From role-ceo
Capture a decision in the team's standard decision-log format and publish it to Notion. Drafts only — publishes on explicit approval. Use when the user says "this is a decision," after a decision-making meeting, or when writing up something irreversible.
npx claudepluginhub sitloboi2012/team-marketplace --plugin role-ceoThis skill uses the workspace's default tool permissions.
**Invocation: user only.** Decision log entries are public artifacts — always confirm before publishing.
Guides Next.js Cache Components and Partial Prerendering (PPR): 'use cache' directives, cacheLife(), cacheTag(), revalidateTag() for caching, invalidation, static/dynamic optimization. Auto-activates on cacheComponents: true.
Guides building MCP servers enabling LLMs to interact with external services via tools. Covers best practices, TypeScript/Node (MCP SDK), Python (FastMCP).
Share bugs, ideas, or general feedback.
Invocation: user only. Decision log entries are public artifacts — always confirm before publishing.
The format matches team-core/CLAUDE.md so anyone on the team can scan a decision in 30 seconds and know what we decided, why, and what would make us reconsider.
From $ARGUMENTS plus any recent context, identify:
If the user just says "log this decision" without context, pull recent Slack threads, Fathom meeting summaries, and Notion pages referencing the topic to reconstruct context. Confirm the summary with the user before proceeding.
Exact format (matches CLAUDE.md — don't improvise):
## <one-line title in decisive voice>
- **Date:** YYYY-MM-DD
- **Owner:** <person>
- **Context:** <2-3 sentences. What forced the decision. Include the evidence or pressure that made this necessary.>
- **Decision:** <what we decided, stated as a call, not a recommendation>
- **Alternatives considered:** <bullets — each with a one-line reason it was rejected>
- **Revisit by:** <a date OR a trigger condition OR "not planned">
Title voice: "Move primary datastore from MySQL to Postgres" not "Considering a possible migration to Postgres." A decision is a commitment; write it that way.
Context is evidence, not narrative. Bad: "We've been discussing this for a while." Good: "Three of the last five incidents traced to MySQL's JSON column performance under our current query load; Postgres benchmarks at 4× throughput on the same schema."
Alternatives: at least two, with honest rejection reasons. "We considered X, rejected because Y" — not "X was inferior."
Revisit by is either a specific date, a specific trigger ("if we hit 50k concurrent users"), or "not planned" (for truly one-way decisions). "When appropriate" is not a valid answer.
Show the draft. Common edits:
On approval: