PM Master turns project and program management into a governed, repeatable workflow that generates verified planning, execution, monitoring, and closeout artifacts.
The fastest way to install a skill directly from the registry.
npx clawhub@latest install pm-master-2
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
Copy this prompt to OpenClaw to install it automatically.
Help me install pm-master-2 using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Get the raw skill files in a ZIP archive.
PM Master is an enforced project and program management operating system for OpenClaw. It supports waterfall, agile, iteration, and hybrid methodologies while using project.yaml as the single source of truth for project state, deliverables, metrics, governance, and checkpoints.
Unlike advisory-only Openclaw Skills, PM Master produces files, applies lifecycle gates, validates artifacts, dispatches domain experts, and blocks delivery when quality, approval, confidentiality, or consistency requirements are not met. It can generate charters, WBS structures, schedules, RAID logs, status reports, EVM metrics, stage-gate reviews, DOCX documents, and program rollups.
python --version
python -m pip install pyyaml
python scripts/init_project.py "Example Platform Program" --type project --methodology hybrid
python scripts/project_state.py show --project /workspace/example-platform/project.yaml
python scripts/project_state.py migrate --project /workspace/example-platform/project.yaml
python scripts/project_state.py show --project /workspace/example-platform/project.yaml
python scripts/consistency_check.py --project /workspace/example-platform/project.yaml
The skill uses a file-centered, schema-driven model with project.yaml as the authoritative record.
| Data area | Organization and purpose |
|---|---|
project.yaml |
Single source of truth containing project or program identity, methodology, phase, intent, lifecycle checkpoint, artifacts, WBS, metrics, governance, and configuration references. Schema version 2 adds _checkpoint. |
| Four dimensions | type, methodology, phase, and intent determine which references, templates, scripts, and deliverables are required. |
_checkpoint |
Records workflow progress and prevents illegal lifecycle skips. States progress through planning, review, baselined, operational, and closed. |
wbs |
Expert-decomposed work packages with IDs, domains, roles, owners, numeric estimates, dependencies, and definitions of done. Leaf packages must respect control.granularity_threshold, normally 10 person-days. |
artifacts |
Index of generated deliverables and their paths, including charters, schedules, RAID logs, reports, plans, closure documents, and rollups. |
metrics |
Operational and earned-value data such as planned value, earned value, actual cost, CPI, SPI, baseline information, and control timestamps. |
governance |
Stage-gate settings, approver identities, communications audit records, approvals, and lifecycle controls. Self-approval is forbidden. |
config.yaml |
Runtime policies for execution backends, quality gates, stage gates, email, confidentiality scanning, and control cadence. |
<slug>_data.yaml |
Intermediate structured data prepared for each rendered deliverable. |
| Markdown outputs | Template-rendered project documents generated by render.py; final delivery requires consistency_check.py exit 0. |
| DOCX outputs | Optional formal documents generated from validated Markdown by render_docx.py. |
| Dispatch and reports | dispatch.py creates an idempotent expert dispatch plan; sub-agents return contract-based JSON reports validated by subagent_check.py. |
| Program rollups | Program views consolidate sub-project milestones, health, blockers, benefits, and dependencies through scripted rollup tools rather than hand-edited parallel sources. |
| Communication audit | Approved outbound messages are recorded under governance.communications; email remains off by default. |
| Confidentiality results | confidentiality_check.py scans text and binary content for configured sensitive tokens before publication. |
The taxonomy separates project versus program scope, waterfall versus agile or hybrid methodology, lifecycle phase, and user intent. This prevents incompatible templates, such as mixing waterfall WBS and stage gates with agile-only backlog structures.
Loading
Agent Throne lets AI agents gain verifiable visibility by challenging for a public #1 ranking with free gauntlets or USDC bids on Base.

Carrusel del Radar F100K turns a sourced trend signal into a reviewed, rendered, and optionally published Instagram carousel with explicit approval gates.

Context Compressor turns verbose Markdown conversation logs into compact, structured memory summaries while preserving critical decisions, lessons, and follow-up tasks.

Memory Fortress Free gives AI agents durable, structured memory that survives context compression, restarts, interruptions, and long-running projects.

A Node.js research skill that turns public Xiaohongshu and RedNote data into actionable insights for content opportunities, competitor monitoring, KOL screening, and market research.

PaintForge turns plain-language requirements, Mermaid input, or repository evidence into validated, interactive standalone HTML diagrams.








































