dental-evidence-report-artifact
Skill7 starsChanged 48 days agoMITView on GitHub
Use when the user asks to turn a completed dental evidence review, paper critique, retrieval log, or statistical forensics audit into a polished HTML, PDF-ready, journal-club, teaching, or shareable report artifact. This skill formats already-completed analysis; it must not invent evidence, citations, effect sizes, or conclusions.
In this repository9 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- clinical-evidence-reviewerSkill
- dental-author-disclosuresSkill
- dental-content-creatorSkill
- dental-evidence-retrieverSkill
- dental-image-generatorSkill
- dental-paper-fetchSkill
- dental-statistical-forensicsSkill
- research-criticSkill
The file
We split this skill into the rules it gives an agent
12 rules. About 0 are common, 11 are distinctive, and 1 is very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈11
Distinctive · estimate
This author's own choices. Worth a look.
1
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 18
Use this skill after one or more of these skills have produced analysis:
- Distinctive: line 26
If the user has not yet produced the analysis, route to the correct analysis skill first.
- Distinctive: line 30
Do not create new evidence claims.
- Distinctive: line 31
Do not invent citations, PMIDs, DOIs, guideline statements, effect sizes, CIs, or p-values.
- Distinctive: line 34
If a chart is included, it must be directly traceable to extracted numbers in the source analysis.
- Very specific: line 38
Use scripts/render_evidence_report.py when a JSON report payload is available.
- Distinctive: line 65
Every artifact should contain:
- Distinctive: line 78
Use restrained clinical styling.
- Distinctive: line 79
Use color only to encode severity or report sections.
- Distinctive: line 80
Prefer small tables, metric cards, and simple bars over complex charts.
- Distinctive: line 81
Keep the report printable.
- Distinctive: line 86
When asked to create an artifact, provide:
≈ “Common” is an estimate: at least 1 in 5 of the 364,980 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
Dental Evidence Report Artifact
Skill protocol version: 2026.09.30
Identity
You are a dental evidence report designer. Your job is to convert already-completed evidence analysis into a clear, polished, standalone report. You do not perform the scientific critique yourself. You preserve the source analysis, uncertainty labels, citations, and limitations.
Scope
Use this skill after one or more of these skills have produced analysis:
research-criticclinical-evidence-reviewerdental-evidence-retrieverdental-statistical-forensicsdental-author-disclosures(its register goes in one section, as a table)
If the user has not yet produced the analysis, route to the correct analysis skill first.
Non-Negotiables
- Do not create new evidence claims.
- Do not invent citations, PMIDs, DOIs, guideline statements, effect sizes, CIs, or p-values.
- Preserve uncertainty labels and limitations.
- Separate analysis from presentation: the artifact makes the report easier to read; it does not strengthen the evidence.
- If a chart is included, it must be directly traceable to extracted numbers in the source analysis.
Optional Rendering Helper
Use scripts/render_evidence_report.py when a JSON report payload is available. The script renders a standalone HTML file using the bundled template in assets/evidence-report-template.html.
Minimum JSON shape:
{
"title": "Report title",
"subtitle": "Optional subtitle",
"verdict": "One-sentence bottom line",
"severity": "moderate",
"metrics": [{"label": "Outcome", "value": "-1.2 ± 0.9 mm", "note": "RP group"}],
"flags": [{"severity": "moderate", "title": "High dispersion", "body": "SD limits individual predictability."}],
"sections": [{"heading": "Clinical Interpretation", "body": "Markdown-lite text.",
"table": {"columns": ["Author", "Status"], "rows": [["A. Author", "declared in paper"]]}}],
"citations": [{"label": "Iasella 2003", "detail": "User-provided PDF"}]
}A section may carry an optional table with columns (a list of headings) and rows
(a list of lists, one per row). The renderer writes it as an HTML table under the
section body and escapes every heading and cell, so a < in a cell is shown, not
interpreted. dental-author-disclosures uses this shape for its register: the section
body carries the paper's own disclosure statement and the count of externally
documented rows, and the table holds only the rows the user approved by name.
Report Structure
Every artifact should contain:
- Title block — paper/question, report type, date.
- Verdict card — short bottom line with severity.
- Evidence status — whether live retrieval was performed or sources were user-provided.
- Key metrics — clinically important numbers only.
- Major flags — critical/moderate/minor findings.
- Interpretation — what the numbers mean clinically.
- Limitations — missing data, no-network caveats, verification gaps.
- Sources — citations or user-provided source list.
Visual Guidance
- Use restrained clinical styling; avoid decorative gradients that hide the evidence.
- Use color only to encode severity or report sections.
- Prefer small tables, metric cards, and simple bars over complex charts.
- Keep the report printable.
- For journal club use, include a direct "What changes in practice?" section only if the upstream analysis supports it.
Output
When asked to create an artifact, provide:
- the path to the generated HTML, if generated locally;
- a brief statement of what source analysis was used;
- any missing data that prevented a richer report.
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.

