---
name: manage-research-portfolio
description: Build and maintain a PI-level command view of Bowen Li's grants, manuscripts, experiments, trainees, collaborations, talks, deadlines, decisions, and risks. Use for weekly planning, deadline radar, project prioritization, workload balancing, status reporting, and converting scattered files or notes into an actionable research portfolio.
---

# Manage research portfolio

1. Gather current evidence from `00_Active_Now`, recent files, calendar or email when authorized, and existing trackers.
2. Create one row per outcome-bearing project, not one row per file.
3. Record objective, current stage, next irreversible decision, next action, owner, deadline, dependency, risk, confidence, and source evidence.
4. Distinguish hard external deadlines from internal targets and undated aspirations.
5. Rank work by scientific or strategic value, deadline pressure, dependency leverage, and PI attention required.
6. Produce a weekly command brief: wins, deadlines in 14/30/90 days, blocked items, decisions only Bowen can make, delegated actions, and projects to pause.
7. Never infer a commitment from an early draft or old file. Mark stale or conflicting evidence.
8. Maintain a decision log so the same issue is not repeatedly reopened without new evidence.

Read `references/portfolio-schema.md` before creating or revising the tracker.