Roadmapping and prioritisation skills
Roadmapping and Prioritisation Skills-md for AI Agents
Roadmapping and prioritisation skills-md are SKILL.md files that teach Claude, Codex and other AI agents to decide what a product should build next and how to sequence it, using the real frameworks and honest roadmap formats rather than a plausible-looking guess. Install one and the agent follows the method a careful product manager would, and shows its working.
6 of 6 skills published, version 1.0.0.
Where to start: begin with prioritisation-frameworks to decide what is worth building, then product-roadmapping to sequence it. For discovery, opportunity-solution-trees picks the customer problem worth solving from real interviews, user-story-mapping slices the journey into releases, roadmap-sequencing puts the ranked work in a buildable order, and delivery-forecasting turns it into dates with a confidence range.
Prioritisation frameworks
Chooses the right prioritisation framework for the decision and applies it with the real scales, not improvised ones. RICE, ICE, WSJF, Kano, weighted and opportunity scoring, MoSCoW and value-versus-effort, with the thresholds that catch a gamed or false-precision score, a protected lane for must-do work, and a sensitivity check before any order is published.
Product roadmapping
Builds a roadmap that survives change: outcomes rather than feature lists, Now-Next-Later confidence horizons, dates only where they are genuinely fixed, and a format matched to the audience. Covers the outcome statement and swap test, the Now cap, the portfolio mix by stage, a feature-factory diagnostic, and a script for when leadership demands dates.
Opportunity solution trees
Connects one product outcome to the customer opportunities that drive it, then to solutions and the tests that prove them, the Teresa Torres way. Covers framing opportunities in the customer's words, the solution-in-disguise test, comparing only sibling opportunities, exploring at least five before going deep, three solutions per target, and assumption tests with a pass threshold and a clear go or no-go.
User story mapping
Lays a product out as the story of someone using it, then slices it into releases, the Jeff Patton way. Covers the backbone of user steps read left to right, details hung by priority, the mile-wide-inch-deep first pass, the walking skeleton, and release lines that each name a target user and a measurable outcome rather than being filled to capacity.
Roadmap sequencing
Turns a ranked backlog into a buildable order: prerequisites first, then the highest cost of delay per size among the work that can actually start, with the riskiest assumptions tested early and long dependency chains split or decoupled. Builds the dependency graph, finds the critical path, caps work in progress with slack, and schedules the fixed dates backwards, so value arrives early and no team sits idle waiting on an unbuilt prerequisite.
Delivery forecasting
Forecasts when roadmap work will land as a date range with a probability, from the team's real throughput, not a single guessed date. Runs a Monte Carlo simulation over completed-item history, reads the 85th percentile for stakeholder dates and the 50th for internal planning, models scope as a low-to-high range, falls back to reference-class forecasting when data is thin, and handles the conversation when a date has already been promised.
No skill matches that search. Clear it or choose All.
What roadmapping and prioritisation skills do for an AI agent
A general model can name a prioritisation framework or sketch a roadmap, but left alone it does the two things that discredit both: it invents a RICE score to three decimals from numbers it guessed, and it draws a tidy timeline the whole company then reads as a promise. These skills stop that. Prioritisation-frameworks carries the real scales and the thresholds that catch a gamed or false-precision score; product-roadmapping keeps confidence and dates on separate axes so the plan survives contact with reality. Each hands back something a PM can defend, not just a number or a chart.
Where to start
Begin with prioritisation-frameworks. Deciding what is worth building comes first, and the skill picks the right framework for your situation, applies it with the real scale, and runs a sensitivity check before it commits to an order. Then move to product-roadmapping to sequence that order across now-next-later horizons and communicate it without turning every date into a commitment you cannot keep. The two compose: one ranks, the other sequences and tells the story. Two more sit alongside them: opportunity-solution-trees, for choosing which customer problem to solve, and user-story-mapping, for slicing that work into releases around a user and a measurable outcome.
What the agent needs from you
Give the agent the candidate work described as problems or outcomes rather than feature names, the strategy or goal it should serve, and whatever real evidence you have for reach and impact. Say who signs off on effort and roughly how much capacity exists. For a roadmap, say who will read it and which dates, if any, are genuinely fixed. Where you have no evidence, say so: the skill marks the input a guess rather than invent a number, and tells you what would change the answer.
Frequently asked questions
Which prioritisation framework should I use?
It depends on the decision, and the prioritisation-frameworks skill picks for you: RICE for a mixed backlog with some evidence, ICE for a fast low-rigour call, WSJF for engineering-led work where urgency and size vary, Kano to tell delight from mere expectation, MoSCoW to communicate what is in and what is cut. It also tells you when a framework is the wrong tool and the real question is strategy.
Do these replace a product manager?
No. They give an agent the method a good PM uses and the discipline to show its working, but a human owns the decision and any commitment to a date or a customer. The skills support a decision, they do not make it for you.
Can I use them with Jira, Linear or a spreadsheet?
Yes. Both skills are tool-agnostic. They tell the agent to enter the scored inputs and the roadmap outcomes into whatever you already use, Jira, Linear, Productboard or a sheet, and run the checks by hand rather than needing an API or plugin.
How many skills are in this set?
All six are published: prioritisation-frameworks, product-roadmapping, opportunity-solution-trees, user-story-mapping, roadmap-sequencing and delivery-forecasting. Download the whole set as one folder from the button above, or take any skill on its own.
Sources and more product management skills
Each skill lists its own sources with the date it was last checked, from the frameworks’ originators, Intercom on RICE, Scaled Agile on WSJF, the Kano model and the DSDM consortium on MoSCoW, and ProdPad, SVPG and Itamar Gilad on roadmaps, through to the practitioners who have run them at scale. Open either skill to read them in full. For the rest of the toolkit, see the Product Management category.
