recommendation-canvas
by deanpetersrecommendation-canvas is a structured AI product recommendation skill for Product Management teams. It helps you evaluate business outcomes, customer outcomes, problem framing, solution hypotheses, positioning, assumptions, and risks before you commit to building an AI feature or product. Use the recommendation-canvas guide to turn a rough idea into a defensible recommendation.
This skill scores 76/100, which means it is a solid listing candidate for directory users who want a structured way to evaluate AI product proposals. The repository provides enough concrete workflow content and template guidance to justify installation, though users should expect a strategy-oriented skill rather than a narrowly automated one.
- Clear trigger and purpose: it is explicitly for evaluating and proposing AI product solutions when deciding whether they deserve investment or recommendation.
- Strong operational structure: the SKILL.md defines core canvas components, and the template provides a ready-to-use recommendation canvas format.
- Good install decision value: the repo includes an example canvas and non-placeholder content, making it easier for agents to understand the expected output quickly.
- No install command, scripts, or support files are provided, so adoption relies on reading the markdown skill and template rather than executing a packaged workflow.
- The content is strategic and framework-heavy, so it may be less helpful for fast tactical tasks or for agents needing highly prescriptive step-by-step automation.
Overview of recommendation-canvas skill
recommendation-canvas is a structured decision-making skill for evaluating AI product ideas before you commit to building them. It helps you turn a vague proposal into a defensible recommendation by mapping business outcome, customer outcome, problem framing, solution hypothesis, positioning, and risk.
What recommendation-canvas is for
Use the recommendation-canvas skill when you need to decide whether an AI feature, workflow, or product is worth investment. It is especially useful for Product Management teams that need a recommendation canvas they can share with stakeholders, not just a brainstorm.
Who should use it
This skill fits PMs, founders, product strategists, and AI solution owners who need to justify an idea with evidence and assumptions. It is less useful if you already have a fully specified feature brief and only need implementation details.
What makes it different
Unlike a generic prompt, recommendation-canvas forces a balanced view: value, uncertainty, and risk. That makes it stronger for recommendation-canvas for Product Management because it helps you explain why the idea matters, what must be true, and how you will know early whether it deserves more work.
How to Use recommendation-canvas skill
Install and load it
Use the repository install flow from the skill file, then start with skills/recommendation-canvas/SKILL.md. If you are browsing manually, also open template.md and examples/sample.md so you can see the target format before drafting your own canvas.
Give the skill a decision, not a theme
The recommendation-canvas install works best when your input is a concrete product choice, such as “Should we add AI summaries to support tickets?” rather than “Explore AI for support.” The skill needs a target user, a business goal, and the context of the decision.
Turn a rough idea into a strong prompt
A weak brief asks for “a recommendation canvas for an AI idea.” A stronger recommendation-canvas usage prompt says who the product is for, what outcome matters, what alternatives exist, and what risk is highest.
Example input shape:
- Product or feature name
- Target persona
- Desired business metric
- Customer pain or job-to-be-done
- Known constraints, risks, or unknowns
- Preferred competitor or current workaround
Read these files first
Start with SKILL.md for the framework, then template.md for the output structure, then examples/sample.md to understand the level of specificity expected. Those three files give you the fastest recommendation-canvas guide to the skill’s logic and formatting.
recommendation-canvas skill FAQ
Is recommendation-canvas just a strategy template?
No. The recommendation-canvas skill is a decision tool for AI product proposals. It is designed to make assumptions visible, not to produce polished marketing copy or a feature spec.
When should I not use it?
Do not use recommendation-canvas when you only need a lightweight idea dump, a technical design doc, or a roadmap ticket. It is strongest when the choice is consequential and you need a recommendation that can survive stakeholder review.
Is it beginner-friendly?
Yes, if you can describe the product idea in plain language. The main challenge is not writing skill; it is giving the canvas enough context to separate business value, customer value, and risk.
How does it fit into an AI product workflow?
It fits early in discovery, before solution design and implementation planning. Teams often use it after initial ideation and before experiments, because the canvas helps decide which assumptions are worth testing first.
How to Improve recommendation-canvas skill
Supply sharper inputs
The best recommendation-canvas results come from specific constraints: target segment, current workaround, success metric, and decision deadline. If you only provide a broad theme, the output will overgeneralize and miss the tradeoffs that matter.
Ask for explicit assumptions and risks
Tell the skill what is uncertain: user trust, data quality, workflow fit, legal exposure, or adoption friction. This improves the recommendation-canvas guide output because the canvas can separate what is proven from what still needs discovery.
Iterate from recommendation to experiment
After the first canvas, ask for a tighter version focused on one audience, one outcome, or one risk. Then request experiments, proof-of-life signals, or positioning alternatives so the recommendation becomes testable instead of abstract.