html-plan is a specialist skill for plans people need to inspect and act on. It keeps the user’s scope, ordering, commitments and terminology, and improves hierarchy and structure without inflating an implementation plan into a strategy document.
The agent decides what the plan needs, such as phases, owners, dependencies, risks and acceptance checks, and adds nothing the source does not support. It delivers one responsive, accessible HTML file and notes any structural interpretation it introduced.
What You Get
- One self-contained HTML plan with semantic headings, lists, tables and landmarks
- Source commitments kept recognizable, with accepted decisions apart from assumptions and open questions
- Sequence and dependency shown only where they exist in the source
- A check that no commitment was dropped and stages stay in the intended order
Common Use Cases
- Rendering an implementation plan so reviewers can see phases and owners at a glance
- Presenting a rollout with risks and acceptance checks
- Cleaning up a roadmap draft without changing what it promises
- Giving a team a readable version of a long planning document
What Makes It Unique
It treats source material as binding. The skill forbids adding a timeline, progress percentage, status badge or dashboard summary unless the source supports it, and it keeps long prose readable without forcing every paragraph into a card. When design-artifact is installed, creative direction is allowed but cannot override traceability.
Under The Hood
The skill is a single instruction file with no scripts. The agent reads your plan and project files and writes one HTML file to your workspace, then inspects it in a browser at wide and narrow widths to check ordering, ownership and overflow. The instructions avoid external services and do not call the network.