Editorial reviewer for tessl.io blog articles. Scores drafts across six strategic dimensions (audience, AEO/GEO, SEO, Tessl alignment, technical depth, structure) with atom-level recommendations grouped under each dimension. Paste-ready output for Tessl's article review page.
78
90%
Does it follow best practices?
Impact
100%
1.66xAverage score across 1 eval scenario
Passed
No known issues
{
"context": "Tests whether the agent reviews a deliberately flawed Trend Analysis blog draft using Tessl's editorial standards. The draft lacks question-shaped headings, buries answers, names only one tool, has no FAQ block, no comparison table, and no internal links. The review must produce a committed single keyword, a written meta description and URL slug, and specific AEO/GEO recommendations referencing actual article content.",
"type": "weighted_checklist",
"checklist": [
{
"name": "Single keyword committed",
"description": "The SEO package commits to exactly ONE primary keyword — does not say 'you could use X or Y' or present multiple options",
"max_score": 10
},
{
"name": "Meta description written",
"description": "The SEO package includes a complete, written meta description of 155 characters or fewer — not delegated to the author with a placeholder",
"max_score": 10
},
{
"name": "URL slug provided",
"description": "The SEO package includes a specific URL slug (e.g., /blog/ai-agent-evaluation) — not left blank or described abstractly",
"max_score": 8
},
{
"name": "AEO/GEO hard fail flagged",
"description": "The review explicitly calls out a Hard Fail or flags that the AEO/GEO score is ≤1 (piece does not ship), given the article's lack of front-loaded answers, no named entities beyond generic descriptions, and buried structure",
"max_score": 10
},
{
"name": "No question-shaped heading flagged",
"description": "The review specifically flags that the article has no question-shaped H2 or H3 heading",
"max_score": 8
},
{
"name": "Low named-entity count flagged",
"description": "The review flags that the article names fewer than 3 specific tools, companies, reports, or products (below the 3–5+ threshold), and suggests specific named entities to add",
"max_score": 8
},
{
"name": "FAQ block recommended",
"description": "The review recommends adding a FAQ block, appropriate for Trend Analysis pieces",
"max_score": 8
},
{
"name": "H2 before/after rewrites",
"description": "The SEO package includes at least one before → after rewrite of an existing H2 heading from the article to carry a searchable keyword term",
"max_score": 8
},
{
"name": "Keyword from priority cluster",
"description": "The committed primary keyword belongs to one of the five named priority clusters (Agent Skills, Context Engineering, AI Agent Evaluation, AI Agent Reliability, or Context Governance), OR is flagged as outside them with a long-tail alternative",
"max_score": 8
},
{
"name": "No internal links flagged",
"description": "The review flags that the article contains no internal links to tessl.io articles",
"max_score": 8
},
{
"name": "Comparison table recommended",
"description": "The review recommends adding at least one comparison table or enumerable list to support AEO/GEO extractability",
"max_score": 8
},
{
"name": "Atom-level recommendations",
"description": "At least two recommendations quote a specific phrase or exact heading text from the article draft rather than offering generic advice (e.g., references 'The Changing Landscape' or 'flying blind' rather than 'improve your opening')",
"max_score": 6
}
]
}