Skip to main content

Introduction

PM Workspace Kit is an opinionated template for the kind of work that happens before production code: product discovery, migration planning, cross-team reviews, and keeping all of it traceable after the initial excitement wears off.

It was extracted from a real 18-month ERP migration planning effort. The domain content (product entities, module playbooks, specific SLOs) was stripped out. What's left is the structural scaffolding that worked — the scripts that enforce it in CI so it doesn't rot, plus a CLI (pmk) and a Slack gateway that turn each PM verb into a repeatable conversation against your real codebase.

What you get​

The kit ships three product surfaces backed by one set of templates and one traceability core.

1. Templates + scripts that enforce discipline​

  • traceability.js — scans markdown front-matter, validates required fields, emits a Mermaid dependency graph + reverse-lookup + orphan report.
  • confluence-sync.js — pulls comments and status labels from Confluence back into your Git-of-record so reviewer activity doesn't evaporate in a wiki.
  • ADR template (MADR-style) with slots for Alternatives Considered and Consequences (positive / negative / neutral).
  • Module migration playbook — 12 sections including Stage 0–4 Strangler Fig checklist and quantitative exit criteria.
  • North-star architecture outline — 10-chapter scaffold (Exec Summary → As-Is → Vision → To-Be → Three Pillars → Roadmap → Migration → Cross-cutting → ADR Index → Appendix).
  • PR / Code review / Runbook / Dashboard / Readiness — five handoff documents so engineering inherits a plan, not prose.

2. pmk CLI — PM verbs as structured conversations​

pmk turns each part of the PM workflow into a named verb. Every verb is a guided LLM session that produces a tracked artifact in your repo:

  • pmk propose → PRD interview → markdown file in docs/prds/
  • pmk ingest mra:--all → loads PKB summaries for every repo in your mra workspace
  • pmk discuss / ask / debug → grounded brainstorming, RAG queries, hypothesis-driven debugging
  • pmk case open prod-checkout-503 → long-lived bug investigation file (state persists across sessions)
  • pmk apply <plan> → walk a decomposed plan task-by-task with explicit confirmation
  • pmk gateway → run a Slack bridge so other PMs and stakeholders can DM pmk from the messengers they already use; channel @mentions create case files

3. Methodology ADRs (pre-written)​

  • Strangler Fig Protocol — four named stages with exit thresholds and rollback playbooks.
  • Dev Harness convention — how to layer Claude Code skills / hooks / subagents across a monorepo.
  • Product Decision Log — a dedicated ADR class for product-facing decisions (pricing, entitlement, MVP scope) so they stop getting re-litigated.
  • pmk × mra bridge — why the CLI delegates code intelligence to multi-repo-agent instead of growing its own grep.
  • pmk gateway (Slack) — why the messenger surface is a host-run bridge, not a SaaS bot.

What you don't get​

  • A project management tool. Tickets still live in Linear / Jira / whatever.
  • A wiki replacement. This kit works with Confluence; it doesn't try to replace it.
  • Auto-generated architecture. Humans still decide, ADRs still get written, migrations still get planned — the kit just gives those acts a shared shape.
  • A hosted SaaS. The CLI and gateway run on your machine against your repos. Nothing leaves the host.

Who it's for​

  • PMs and systems analysts in companies that have outgrown free-form Notion docs.
  • Engineering leads about to plan a platform migration and dreading the 50-page design doc.
  • Staff+ engineers who want a lightweight ADR + Strangler Fig toolkit they can drop into any repo.
  • Cross-functional teams that want stakeholders to interact with PM artifacts (PRDs, cases, decisions) over Slack instead of "please open this Confluence page".

Next​