Claude Code Should Output HTML, Not Just Markdown
Still asking Claude Code for long Markdown dumps? Thariq says complex plans, code reviews, and research reports work better as clickable, filterable, reusable HTML workspaces.
Published 143 days ago. Content may be outdated.
Thariq Shihipar published a short but useful essay on X: Using Claude Code: The Unreasonable Effectiveness of HTML. The point is not a hidden command or a new model trick. It is simpler: ask Claude Code to output HTML.
That sounds small, but it changes the shape of the work. Many people still ask coding agents for Markdown, then manually read, sort, copy, annotate, and reshape the result. Thariq’s argument is sharper: if an output is meant to be reviewed, clicked, compared, edited, or shared, HTML is often closer to the real workspace.
The Core Idea
Claude Code should not only be treated as a code generator. It can also produce temporary dashboards, review pages, planning boards, reports, prototypes, and lightweight editors.
Markdown is great for notes. It is weak when the work becomes dense. Plans, code reviews, research summaries, issue triage, and product comparisons quickly need hierarchy, visual emphasis, filtering, status, and interaction.
HTML gives you those pieces in one local file. Cards, tables, collapsible sections, progress bars, buttons, filters, and inline editing can turn a model response into something closer to a working interface.
Markdown Feels Like an Endpoint
The biggest strength of Markdown is simplicity. Its weakness is the same thing.
When Claude Code outputs a Markdown plan, you mostly read it. If you want to compare options, you scroll. If you want to mark risks, you copy the content somewhere else. If you want to share it with a team, you often reshape the format.
That is why many AI outputs feel useful but never become used. The content exists, but it is not operational.
HTML can turn the same content into a page you can scan, filter, revisit, and act on.
From Text Generation to Tool Generation
The valuable shift is not aesthetic. It is functional.
Instead of asking whether Claude Code can write a function, fix a bug, or explain a file, ask whether it can build a temporary workspace around the task.
For example:
- Generate a code review page with severity levels and file paths
- Turn a backlog into a filterable priority board
- Compare multiple implementation plans side by side
- Produce a clickable prototype for a feature idea
- Convert research notes into an expandable, searchable report
These outputs do not need to ship to production. They only need to reduce the cost of understanding and acting on the result.
Why This Matters for Agents
The hard part of agentic work is not only execution. It is handoff.
A model may finish a task, but a human still has to judge risks, split follow-ups, share context, record decisions, and decide what happens next. If that handoff remains a long Markdown block, it becomes another pile of text to manage.
HTML works as a bridge. It is lighter than building a full app, but richer than static notes. It lets one agent run leave behind something usable.
The deeper lesson: do not only optimize prompts. Optimize the artifact the agent gives back.
HTML Is Not Always Worth It
There is a tradeoff.
HTML is heavier than Markdown. It can create unnecessary styling, it needs browser preview, and it can overcomplicate simple tasks. If you only need five commands or a quick summary, Markdown is still faster.
A practical rule: ask for HTML when the output needs repeated reading, team review, status tracking, comparison, filtering, or follow-up editing.
For disposable notes, stay with Markdown.
A Prompt You Can Reuse
Try adding this to your next complex Claude Code task:
Do not only output Markdown. Generate a single-file HTML report with an executive summary, risk levels, collapsible details, action items, and next-step recommendations. Keep all styles inline so I can open it locally.For code review, make it more specific:
Output the review as an HTML page. Show each issue as a card with severity, file path, impact, suggested fix, and testing notes. Add top navigation for filtering by severity.The goal is not decoration. The goal is to make the output usable after the model stops talking.
Takeaway
The useful distinction is simple: Markdown is a report. HTML is an interface.
Reports are for reading. Interfaces are for continuing the work.
If you already use Claude Code but only ask it for Markdown, you may be leaving a lot of value on the table. The next step is not always a better prompt. Sometimes it is a better output format.
Sources:
More Articles
