render-report
Skill18 starsChanged 4 months agoMITView on GitHub
Render existing .ux/audits reports into self-contained, offline HTML pages — findings as cards, the captured screenshots as an interactive walk-through flipbook, plus the roll-up dashboard and an index landing page. Use for "render the report to HTML", "HTML view of the audit", "open the audit in a browser", "flipbook walkthrough", "make a shareable report page".
In this repository6 other files it gives its agents.
- AGENTS.mdAGENTS.md
- docs/agents.mdAGENTS.md
- audit-cujSkill
- spec-cujSkill
- usability-auditSkill
- ux-auditSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 45
Run the transform over the selected reports, then refresh the index:
- Very specific: line 76
Derived view only: Never edit the report Markdown, the report contract, or host application code.
- Very specific: line 85
Every selected report has a self-contained .html companion beside it, and index.html is refreshed.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Render report — a self-contained HTML view of an audit
Turn the Markdown reports under a host's .ux/audits/ into self-contained HTML you can
open in a browser or share: findings as priority-badged cards, the live screenshots as an
interactive walk-through flipbook, the roll-up as a priority dashboard, and an
index.html landing page. The transform is
scripts/render_report_html.py.
The Markdown report is the source of truth; the HTML is a derived view. This skill never
edits a report's Markdown, the shared report contract,
or any host application code — it only writes .html companions beside the .md files, under
.ux/audits/. Running it changes no finding, count, or verdict.
This is the on-demand path. The audit skills already render the HTML as their final step; use this to (re)render reports that already exist — after pulling a branch, to refresh the view, or to render an older run.
Inputs
$ARGUMENTS:
target— which reports to render. A single report (.ux/audits/cuj-<ts>.md), a glob (.ux/audits/*.md), or omitted to render every report in the host's.ux/audits/.--index— also (re)generate theindex.htmllanding page fromindex.md. Implied when rendering the whole directory.
Workflow
1. Locate the reports
Find the Markdown reports under the host's .ux/audits/. If the directory is absent or holds
no *.md reports, stop and say so — there is nothing to render; point the user at
/usability-audit, /cuj-audit, or /ux-audit to produce one first. Do not create reports
here; this skill only renders what already exists.
Exit criteria: a non-empty set of report paths, or a clean stop with the reason.
2. Render each report, and the index
Run the transform over the selected reports, then refresh the index:
python3 scripts/render_report_html.py .ux/audits/<report>.md # or .ux/audits/*.md
python3 scripts/render_report_html.py --index .ux/audits/index.mdEach <report>.md yields <report>.html beside it; rollup-*.md renders as the dashboard;
index.md renders as the landing page. The output is deterministic — re-rendering an
unchanged report produces a byte-identical .html, so there is no diff churn.
Exit criteria: one .html per selected report plus index.html, each self-contained
(images inlined as data: URIs, no external references — it opens offline).
3. Report what was written, and confirm the footprint
List the .html files written and confirm every change is confined to .ux/audits/:
python3 scripts/audit_safety.py <host-repo>This exits 0 only when both halves hold — nothing outside .ux/audits/, and the .html
files you just wrote observed inside it. A render that produced no files exits 1, which is
the correct answer: there was nothing to render or nothing got written.
Exit criteria: audit_safety.py exits 0 on both halves: nothing changed outside
.ux/audits/, and the .html files this run wrote inside it were observed and listed.
Boundaries
- Derived view only. Never edit the report Markdown, the report contract, or host
application code. The
.htmlis generated from the.md; the Markdown stays canonical. - Never write outside
.ux/audits/. The.html, the dashboard, andindex.htmlall live there, so the suite-wide safety invariant is unaffected. - Never fabricate. Render what the report says; add no finding, count, image, or verdict the Markdown does not carry.
Exit criteria (done when)
- Every selected report has a self-contained
.htmlcompanion beside it, andindex.htmlis refreshed; each opens offline with no external references. - An absent/empty
.ux/audits/stopped the run with a reason instead of inventing a report. audit_safety.pyexits 0 on both halves: nothing changed outside.ux/audits/, and the writes this run made inside it were observed and listed. (git statusalone is not the check — once.ux/audits/is gitignored it shows nothing, which a run that wrote nothing satisfies identically.)
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

