Sign inGet started
Skills / Project_300

Scaled Agile Framework (SAFe)

Fund value streams, not projects.

SAFe is the world's most widely adopted framework for scaling Agile across large organizations. Developed by Dean Leffingwell, it organizes teams of 50–125 people into Agile Release Trains aligned to value streams, operating on synchronized 10-week Program Increments with PI Planning, WSJF prioritization, and Lean Portfolio Management. This skill is built for practitioners already working inside SAFe — Release Train Engineers, Product Managers, Agile coaches, and portfolio leaders who need a working operating model, not introductory Agile concepts.

By Dean Leffingwell · Free
Specimen 01 · Live diagnosisScaled Agile Framework (SAFe)
Input

“We have our first PI Planning in 3 weeks. We have 6 teams, about 70 people total. Nobody knows what's supposed to happen on Day 1 and our RTE is…”

Diagnosis
I work from the actual SAFe framework: PI Planning agendas, WSJF formulas, ART launch sequencing, Lean Portfolio governance.
Full transcript ↓
Calibrated referenceagent-skills.ai
The gap

Turn 50 teams into one ART — run PI Planning, prioritize with WSJF, and fund value streams instead of projects

SAFe organizes large organizations into Agile Release Trains (ARTs) — stable teams of 50–125 people aligned to a shared value stream. ARTs operate on a 10-week Program Increment (PI) cadence: five 2-week development sprints plus one Innovation & Planning (IP) sprint. Every PI begins with PI Planning — a 2-day big-room event where all teams jointly build PI Objectives, map cross-team dependencies on the Program Board, and ROAM risks as Resolved, Owned, Accepted, or Mitigated. Prioritization across the backlog uses WSJF (Weighted Shortest Job First: Cost of Delay ÷ Job Size) to rank Features and Epics by business value delivered per unit of time invested. At the portfolio level, Lean Portfolio Management (LPM) replaces project-based funding with value stream budgets, governed by a Portfolio Kanban with Epic hypothesis statements and Lean Budget guardrails. The PI closes with an Inspect & Adapt event combining a System Demo, PI metrics review, and a structured Problem-Solving Workshop using fishbone diagrams and an improvement backlog.

The problem

Most organizations succeed with Agile at the team level — sprints complete, stories ship, Scrum ceremonies run. The crisis hits at scale: teams can't align on cross-team dependencies, product roadmaps have no enterprise prioritization logic, and annual project-based funding destroys the flow that Agile teams worked to achieve. SAFe exists to solve the coordination and governance failures that Scrum was never designed to handle.

The solution

Apply the most prescriptive and actionable parts of SAFe — PI Planning facilitation, WSJF prioritization, Program Board ROAM, and Lean Budget design — with a coach who knows the exact roles, timeboxes, artifacts, and failure modes of every ceremony.

You bring
  • Your ART composition — number of teams, team names, and headcount
  • A list of Features or Epics you need to prioritize (for WSJF analysis)
  • Your current organizational structure — value streams, portfolios, and funding model
  • Specific SAFe ceremony challenges — e.g., a PI Planning that lost momentum, an Inspect & Adapt that produced no improvement backlog, or a Program Board with unROAMed risks
You get
  • A PI Planning agenda with team breakout scripts, timebox guides, and facilitation interventions for the most common failure modes
  • A WSJF-ranked Feature list with Cost of Delay component scores, job-size estimates, and a prioritization rationale you can present to stakeholders
  • An ART Launch Readiness scorecard across organizational, team, and leadership dimensions with a sequenced preparation checklist
  • A Lean Portfolio Management design with Portfolio Kanban state definitions, Epic hypothesis statement templates, and Lean Budget guardrail recommendations
In action

Watch the methodology work.

Three specimens from a single real session: the same situation, unaided and calibrated, the full transcript, and the skill answering live in the channel where the work happens.

Fig.01The same situation, two ways
Specimens A · B
Specimen AVague
Unaided

“Your 6-team ART is 3 weeks from its first PI Planning. The Program Backlog has 200 items with no prioritization logic. Product Managers don't know which Features to present or how to defend their choices. The RTE has never facilitated a 2-day big-room event and isn't sure what 'team breakouts' actually means in practice. Leadership is asking what the tangible output of PI Planning will be and whether it's worth taking 70 people offline for 2 days.”

×Vague — nothing concrete to act on.
Specimen BCalibrated
With Scaled Agile Framework (SAFe)

“You walk into PI Planning with 15 WSJF-scored Features ready for team consumption, a scripted 2-day agenda with timebox calls, and a Program Board pre-loaded with team swim lanes. Day 1 produces draft PI Objectives from all 6 teams and a Program Board showing every cross-team dependency with ROAM status. Day 2 closes with committed PI Objectives, a fully ROAMed risk board, and a confidence vote of 4.1/5. Leadership receives a PI Objectives table with Business Owner-scored business value they can read without knowing what a story point is.”

Specific, grounded, and ends with one clear next step.
Fig.02A real session, unedited
In-app · assistant
Scaled Agile Framework (SAFe)Claude
Scaled Agile Framework (SAFe)
I work from the actual SAFe framework: PI Planning agendas, WSJF formulas, ART launch sequencing, Lean Portfolio governance. What's the most pressing challenge in your SAFe implementation right now? Common starting points: preparing for an upcoming PI Planning event, prioritizing a bloated Program Backlog with WSJF, assessing readiness to launch a new ART, or redesigning your portfolio funding model.
We have our first PI Planning in 3 weeks. We have 6 teams, about 70 people total. Nobody knows what's supposed to happen on Day 1 and our RTE is panicking about facilitating it.
Three weeks is tight but workable. Let's build the Day 1 structure, then flag the pre-PI prep that must happen before the event — because if it doesn't, Day 1 breaks down regardless of the agenda.
Day 1 runs in four sequential blocks:
Morning: Context & Vision (3–4 hours)- Business Context: An executive presents the business state, strategic themes, and PI Themes. Non-optional — teams need to understand why the work matters before planning.- Product/Solution Vision: Product Management walks through the top 10–15 Features planned for this PI. Each Feature needs a description and acceptance criteria.- Architecture Vision: System Architect presents the Architectural Runway and any Enabler Features the ART needs.- Planning process overview: Agenda, team norms, tool orientation.
Afternoon: Team Breakouts, Iteration 1 (3–4 hours)Teams work at their tables: capacity planning, story breakdown per sprint, dependency identification. Any story requiring work from another team gets flagged on the Program Board immediately.
Late Afternoon: Draft Plan ReviewEach team does a 2-minute readout — capacity, stories planned, dependencies flagged, preliminary risks posted.
Evening: Management Review & Problem-SolvingRTE and Product Managers review the draft plans, identify scope adjustments and resource conflicts to resolve before Day 2 begins.
What does your pre-PI preparation look like — specifically the Program Backlog? Are Features refined and estimated, or will teams be seeing them for the first time on Day 1?
Our backlog is a mess — about 200 items with no prioritization. Product Managers aren't ready to present. They don't know which Features to prioritize or how to defend those choices.
200 items is a backlog problem, and you have three weeks to fix the top of it. For PI Planning to function, Product Management needs 10–15 PI-ready Features — each with a clear description, acceptance criteria, rough story-point estimate, and a WSJF score explaining why it made the cut.
You don't need to clean up all 200. You need the top 15–20 and a defensible WSJF rationale for why those are the priority.
Let's run a WSJF pass right now. Give me your top 3–5 Feature titles and a one-line description each. I'll walk you through scoring User/Business Value, Time Criticality, and Risk Reduction/Opportunity Enablement for each one and compute relative WSJF scores — the exact artifact your Product Managers need to present at PI Planning.
Here are three: 1) Mobile checkout redesign — customers are abandoning cart on mobile at high rates. 2) API rate limiting — one customer is hammering our API and degrading service for all others. 3) Reporting dashboard v2 — the product team has been requesting better analytics for 6 months.
Three very different value profiles. Scoring each on the Cost of Delay components using a 1–13 Fibonacci relative scale:
Mobile Checkout Redesign- User/Business Value: 13 — direct revenue impact; cart abandonment is quantifiable lost revenue- Time Criticality: 8 — every week of delay is lost sales; cost compounds- Risk Reduction/OE: 3 — a UX fix, not a risk mitigation- Cost of Delay: 24
API Rate Limiting- User/Business Value: 8 — protects existing revenue, prevents churn- Time Criticality: 13 — active degradation risk; an SLA violation could trigger contract penalties- Risk Reduction/OE: 13 — this is the definition of risk reduction; unmitigated it threatens platform stability for all customers- Cost of Delay: 34
Reporting Dashboard v2- User/Business Value: 5 — internal productivity improvement- Time Criticality: 2 — no hard deadline; backlogged 6 months with no apparent urgency- Risk Reduction/OE: 3 — opportunistic, not urgent- Cost of Delay: 10
Now divide by Job Size. Give me a rough relative estimate: small/medium/large maps to approximately 1/3/8 in Fibonacci units.
API rate limiting is small. Mobile checkout is large. Reporting is medium. This already makes sense — we've been arguing internally about checkout vs. API for two months and WSJF shows immediately that API should go first.
Exactly — and that's WSJF's core job: convert a stakeholder argument about priorities into a conversation about economics.
Final WSJF scores:- API Rate Limiting: 34 ÷ 1 = 34.0 — do this PI, first- Reporting Dashboard v2: 10 ÷ 3 = 3.3- Mobile Checkout Redesign: 24 ÷ 8 = 3.0
Notice Reporting narrowly edges Checkout in WSJF because of job size. Before you accept that ranking, challenge whether Checkout is decomposable. There is almost certainly a Phase 1 — fix only the payment confirmation step — that scores a 3 or 5 on job size instead of 8, which would push its WSJF score above 4.0 and above Reporting.
For PI Planning prep: run your remaining top 12–15 Features through this same scoring exercise with Product Managers next week. Two hours, a shared spreadsheet, and you walk into PI Planning with a ranked Program Backlog backed by economic logic — not the loudest stakeholder in the room.
Shall I build the full WSJF scoring template for your Feature backlog, or do you want to move to the Day 2 agenda and the Program Board ROAM structure?
Reply to Scaled Agile Framework (SAFe)
What the skill is doing
01PI Planning Facilitation
02WSJF Feature Prioritization
03ART Launch Readiness Assessment
Fig.03The same skill, in the channel you already use
WhatsApp · live
9:41
SAScaled Agile Framework (SAFe)typing…
Messages are end-to-end encrypted. No one outside of this chat can read them.
Today
We have our first PI Planning in 3 weeks. We have 6 teams, about 70 people total. Nobody knows what's supposed to happen on Day 1 and our RTE is panic…9:14✓✓
I work from the actual SAFe framework: PI Planning agendas, WSJF formulas, ART launch sequencing, Lean Portfolio governance.9:14
What's the most pressing challenge in your SAFe implementation right now?9:15
Message

The same skill, where the work happens.

No new app to learn. The methodology runs over the WhatsApp Business API, so the answer lands as a reply in the thread you’re already in — same rigour, zero context-switch.

Reads the situation, names the pattern, returns one concrete next move.
Delivered in seconds, inside a conversation that already exists.
Specimen · WhatsApp Business API · live
Capabilities

What it does, specifically.

Each capability is a distinct move drawn straight from the source methodology — not a generic assistant guessing.

CapabilityC-01

PI Planning Facilitation

Walk through the complete 2-day PI Planning event — from pre-PI prep and Business Context presentation through team breakouts, dependency identification, risk ROAMing, and final PI Objective commitment. Get facilitation scripts, timebox guides, and specific interventions for common breakdowns such as teams in 'fake planning,' dependency avoidance, and uncommitted stretch objectives.

Based on SAFe's PI Planning event specification, which defines a mandatory 2-day agenda with Business Context, Product Vision presentations, two-iteration team breakouts, draft and final PI Objectives, and a Program Risks ROAM as required outputs of every ART planning cycle.
CapabilityC-02

WSJF Feature Prioritization

Apply the Weighted Shortest Job First formula — Cost of Delay (User/Business Value + Time Criticality + Risk Reduction/Opportunity Enablement) divided by Job Size — to rank your Program Backlog Features or Portfolio Epics. Work through each cost-of-delay component for specific features and produce a relative priority sequence grounded in business value economics rather than stakeholder opinion or loudest voice.

Based on SAFe's WSJF model, derived from Don Reinertsen's flow economics and implemented in SAFe's Program and Portfolio Kanbans as the standard prioritization method for Features and Epics at every configuration level.
CapabilityC-03

ART Launch Readiness Assessment

Evaluate organizational readiness to launch an Agile Release Train across four dimensions: leadership alignment, value stream clarity, team formation, and tooling/environment readiness. Identify blockers and sequence the preparation steps — from value stream identification through RTE selection and team topology design — to arrive at PI Planning ready.

Based on SAFe's ART Launch Roadmap, which defines a 12-week preparation sequence including value stream identification, ART sizing, coaching preparation, and PI Planning as the launch event that officially activates the train.
CapabilityC-04

Lean Portfolio Management Design

Design or improve your Lean Portfolio Management implementation: define Portfolio Kanban funnel, analysis, backlog, implementing, and done states; create Epic hypothesis statement templates with measurable outcomes; establish Lean Budget guardrails per value stream; and set up the Portfolio Sync cadence and Epic Owner accountability model to replace project-based governance.

Based on SAFe's Lean Portfolio Management competency, which replaces traditional project-based funding with value stream budgets and Epic-level investment governance via Portfolio Kanban, WSJF prioritization, and participatory budgeting cycles.
CapabilityC-05

Program Board Dependency ROAM

Build and work a Program Board — the artifact that makes cross-team dependencies visible across the full PI. Apply the ROAM framework (Resolved, Owned, Accepted, Mitigated) to each identified risk and dependency, ensuring every critical dependency leaves PI Planning with a named owner, a resolution status, and a mitigation action.

Based on SAFe's Program Board artifact and ROAM risk framework, both defined as mandatory outputs of PI Planning for every ART, providing the dependency and risk transparency required to hold PI Objective commitments.
Tested

Graded before it shipped.

Every skill is scored against independent scenarios for methodology fidelity before it goes live — not vibes, a rubric.

What it produces
OutputD-01

PI Objectives Table

Structured team-level and ART-level PI Objectives, each with a business value score (1–10 Fibonacci), committed/stretch designation, and measurable acceptance criteria — the official commitment output of PI Planning that leadership can read without knowing what a story point is.

OutputD-02

WSJF Priority Matrix

A ranked Feature list showing User/Business Value, Time Criticality, Risk Reduction/OE scores, Job Size estimates, and computed WSJF scores — ready to sequence your Program Backlog for the upcoming PI with an economics-based rationale for each prioritization decision.

OutputD-03

ART Launch Readiness Scorecard

A dimension-by-dimension assessment of organizational readiness to launch an Agile Release Train, with red/yellow/green status per dimension, blockers identified, and a sequenced preparation checklist covering leadership alignment, value stream definition, team formation, and tooling.

OutputD-04

Program Board Dependency Map

A sprint-by-sprint dependency visualization across all ART teams, with each inter-team dependency annotated with its ROAM status (Resolved/Owned/Accepted/Mitigated), the accountable team, and the resolution or mitigation action — built during PI Planning and maintained at ART Sync.

The source

Grounded in the original work.

Every answer traces back to a real source and the practitioner who wrote it — not a secondhand summary. Here is the source of record.

Source authorA-01

Dean Leffingwell

Dean Leffingwell is the creator of the Scaled Agile Framework (SAFe) and Chief Methodologist at Scaled Agile, Inc. He authored 'Scaling Software Agility' (2007) and 'Agile Software Requirements' (2011), and has spent decades building the framework documentation, role guides, and certification curriculum that now serve thousands of enterprises globally. SAFe is deployed at more than half of the Fortune 100 and supports one of the largest Agile certification ecosystems in the world, spanning SP, SPC, RTE, PO/PM, and LPM credentials.

Status · Inspired by Dean Leffingwell’s work — not yet claimed. Are you Dean Leffingwell?
Primary sourceS-01

Scaled Agile Framework (SAFe) — scaledagile.com

by Dean Leffingwell

Creator of SAFe; Chief Methodologist at Scaled Agile, Inc.; author of 'Scaling Software Agility' (2007) and 'Agile Software Requirements' (2011); framework deployed at thousands of global enterprises including a majority of the Fortune 100.

Read the original ↗
Citationscaledagile.com
In the build queue

Be first to run it.

Scaled Agile Framework (SAFe) is being built right now. Leave your email and we’ll tell you the moment it goes live.

Notify meEmail
At launchWe have PI Planning in 3 weeks, I'm the RTE, and I need to build the Day 1 and Day 2 agenda, get Product Managers to present a WSJF-prioritized Feature set, and make sure teams can identify dependencies during breakouts. Where do we start?