tools/project-manager/skills/project-manager/SKILL.md
Project management skill for roadmapping, backlog grooming, vision documents, and shipping tracking. Use when the user wants to plan what to work on next, review the roadmap, add/prioritize backlog items, update the vision, log shipped work, or run a retro. Also triggers on "what should I work on", "update the roadmap", "add to backlog", "what have we shipped", "project status", or any planning/prioritization discussion. This skill provides the file structure and conventions; the product-owner agent provides the strategic reasoning.
npx skillsauth add mjmorales/claude-prove project-managerInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Manage the project's strategic planning layer: vision, roadmap, backlog, and ship log. This skill owns the file formats and bootstrap process. For strategic reasoning (prioritization, scope decisions, trade-off analysis), delegate to the product-owner agent.
All planning artifacts live in planning/ at the project root:
planning/
VISION.md # Project pillars, north star, target audience
ROADMAP.md # Milestone-based macro + Now/Next/Later micro
BACKLOG.md # ICE-scored work items, tagged by domain
ship-log.md # Running log of completed work
decisions/ # Consolidated decision references (permanent)
When invoked, determine which operation the user needs and follow that workflow. If unclear, ask.
init — Bootstrap the planning directoryRun this on first use or when planning/ doesn't exist.
planning/ directory$PLUGIN_DIR/tools/project-manager/skills/project-manager/assets/templates/) and populate with project-specific content.prove/archive/ for contextreview — Planning review sessionA conversational check-in on project state. Delegate to the product-owner agent after loading context.
git log --oneline -20 for recent momentum.prove/TASK_PLAN.md and .prove/PROGRESS.md for in-flight workgroom — Backlog groomingAdd, reprioritize, or clean up backlog items.
planning/BACKLOG.mdFor complex prioritization decisions, delegate to the product-owner agent.
ship — Log completed workRecord what was shipped. Run after completing a prove task or merging significant work.
planning/ship-log.md.prove/archive/ for recently archived tasksgit log for recent mergesconsolidate — Consolidate ephemeral decisions into permanent referencesThe .prove/decisions/ directory accumulates individual decision records from brainstorm sessions — dated, per-topic, often overlapping. This operation distills them into clean, organized docs in planning/decisions/.
.prove/decisions/ and scan .prove/archive/*/SUMMARY.md for contextplanning/decisions/<theme>.mdWhen to run: After completing a milestone, after a burst of brainstorming sessions, or when .prove/decisions/ has accumulated enough new records that reference docs feel stale.
Important: This is a rewrite, not an append. Each consolidation re-reads all source material and produces a fresh version of each theme file. The .prove/decisions/ files are NOT deleted — they remain as the raw audit trail.
retro — Lightweight retrospectiveA quick reflection on recent work. No fixed cadence — run when it feels useful.
The vision doc answers: what is this project, who is it for, and what experience/outcome are we creating?
Template: $PLUGIN_DIR/tools/project-manager/skills/project-manager/assets/templates/VISION.md
Key sections:
The roadmap uses milestones as the macro structure and Now/Next/Later as the micro structure within each milestone.
Template: $PLUGIN_DIR/tools/project-manager/skills/project-manager/assets/templates/ROADMAP.md
Milestones are project-specific. During init, work with the user to define milestones that match their project type and goals. Each milestone should have a clear goal and produce something testable/demoable.
Within each milestone (micro), use Now/Next/Later:
The backlog is a single prioritized file with ICE scoring.
Template: $PLUGIN_DIR/tools/project-manager/skills/project-manager/assets/templates/BACKLOG.md
Each item has:
backlog, ready, in-progress, doneItems are sorted by ICE score descending. When an item moves to "in-progress", it feeds into /prove:task-planner for implementation planning.
If the backlog exceeds ~50 items, consider splitting into separate files by domain or milestone.
A running log of completed work, newest first.
Template: $PLUGIN_DIR/tools/project-manager/skills/project-manager/assets/templates/ship-log.md
Each entry has:
.prove/archive/ if applicableRetro sections are interspersed when they happen.
Permanent, theme-organized reference docs distilled from ephemeral .prove/decisions/ records.
Template: $PLUGIN_DIR/tools/project-manager/skills/project-manager/assets/templates/decisions.md
Each theme file contains:
The planning layer sits above the prove execution pipeline:
VISION.md → informs → ROADMAP.md → feeds → BACKLOG.md
↓
pick item, run /prove:task-planner
↓
/prove:orchestrator executes
↓
run /project-manager ship
↓
ship-log.md updated, BACKLOG.md cleaned
This skill provides structure. The product-owner agent provides judgment. The division:
| This skill does | The agent does | |-----------------|----------------| | File formats and templates | Strategic prioritization | | Bootstrap and init | "What should we work on next?" | | Read/write planning files | Trade-off analysis | | Pipeline integration | Scope decisions and pushback | | Ship logging | Teaching PM concepts |
When the user's request requires strategic thinking (not just file operations), spawn the product-owner agent with the relevant planning file contents as context.
tools
Clean and compact prove's durable memory layers — team Lore, the Codex (scrum decisions), annotations, and contributor artifacts — keeping the tribal knowledge that grows team accuracy and folding away the rot. Triggers on "janitor", "clean the lore", "compact the lore", "compact the codex", "memory cleanup", "clean up team memory", "prune stale decisions", "tidy tribal knowledge", "lore cleanup". You are the driver: the CLI emits inventories and executes writes; the `memory-janitor` agent judges each entry; a per-team batch gate approves; nothing is ever deleted — consolidation, promotion, and supersession only.
testing
Anchor session context into prove primitives before compaction and rehydrate from them after. Built-in compaction summarizes by recency and drops the claude-prove state an agent needs to reorient; this skill externalizes volatile context into durable anchors (scrum tasks, decisions, run-state, a compact-anchors pointer file) pre-compact, then runs a deterministic reorientation sequence post-compact. Use before a manual /compact, when context is about to auto-compact, or immediately after a compaction. Triggers on "smart compact", "prepare for compaction", "anchor before compact", "context is getting long", "rehydrate", "reorient after compact".
tools
Apply model-driven CONTENT reshaping to stored run artifacts that sit behind the current schema, on explicit operator invocation only. Triggers on "migrate runs", "migrate run artifacts", "run content migration", "reshape run artifacts", "bring runs to current schema". You are the driver: the `run-state migrate-runs` CLI mechanically detects which artifacts are behind and emits a plan naming each one plus its migration-instruction file; you read the instructions and reshape the prose/findings, gated by the operator. The deterministic `schema migrate` handles structural column moves; this skill covers only the content reshaping beyond them. Never run as a background or resident loop — only when the operator asks.
tools
Synthesize the 7-section risk-forward Review Brief from a run's reasoning log. Triggers on "reasoning brief", "review brief", "synthesize the brief", "generate the brief", "brief the run", "brief for review", "story brief". You are the driver: the `acb brief` CLI renders a mechanical preservation-safe backbone and proves preservation; you synthesize the narrative prose (summary + changes), single-pass or multipass over episode chunks, then gate it through Stage-1 (mechanical, blocking) and Stage-2 (prose judge, advisory).