CtrlK
BlogDocsLog inGet started
Tessl Logo

mcollina/documentation

Creates, structures, and reviews technical documentation following the Diátaxis framework (tutorials, how-to guides, reference, and explanation pages). Use when a user needs to write or reorganize docs, structure a tutorial vs. a how-to guide, build reference docs or API documentation, create explanation pages, choose between Diátaxis documentation types, or improve existing documentation structure. Trigger terms include: documentation structure, Diátaxis, tutorials vs how-to guides, organize docs, user guide, reference docs, technical writing.

91

1.11x
Quality

87%

Does it follow best practices?

Impact

96%

1.11x

Average score across 5 eval scenarios

SecuritybySnyk

Passed

No known issues

Overview
Quality
Evals
Security
Files

criteria.jsonevals/scenario-4/

{
  "context": "Tests whether the agent follows tutorial-specific Diátaxis patterns: verb-led title, Goal→Prerequisites→Numbered steps→Verifiable result→Final outcome structure, minimising explanation in favour of doing, and ensuring every step produces a visible testable result.",
  "type": "weighted_checklist",
  "checklist": [
    {
      "name": "Verb-led title",
      "description": "The document title starts with an imperative verb (e.g., 'Build', 'Create', 'Set up', 'Write', 'Run') rather than a noun or 'How to'",
      "max_score": 10
    },
    {
      "name": "Goal statement present",
      "description": "The document opens with an explicit goal statement telling the reader what they will have achieved by the end (e.g., 'By the end of this tutorial, you will have...')",
      "max_score": 10
    },
    {
      "name": "Prerequisites section",
      "description": "A prerequisites or 'Before you begin' section is present before the first step",
      "max_score": 8
    },
    {
      "name": "Numbered steps",
      "description": "The main body of the tutorial uses a numbered list (ordered list) for the procedural steps",
      "max_score": 8
    },
    {
      "name": "Verifiable result per step",
      "description": "At least 3 individual steps include an explicit expected output or verification check the reader can perform (e.g., showing expected terminal output, telling the reader what they should see)",
      "max_score": 12
    },
    {
      "name": "Final outcome stated",
      "description": "The document ends with a clear final outcome or summary of what has been accomplished",
      "max_score": 8
    },
    {
      "name": "Minimises explanation",
      "description": "The document does NOT contain long conceptual paragraphs explaining why things work the way they do — explanation is limited to one sentence or less per step",
      "max_score": 10
    },
    {
      "name": "Maximises doing",
      "description": "The majority of the document consists of actions to perform (commands to run, files to create) rather than prose descriptions",
      "max_score": 10
    },
    {
      "name": "Beginner self-sufficiency",
      "description": "The tutorial does NOT reference external resources or documentation that a beginner would need to consult in order to complete any step",
      "max_score": 8
    },
    {
      "name": "No reference table mixing",
      "description": "The document does NOT include a table of all options, flags, or parameters (reference material) mixed into the tutorial body",
      "max_score": 8
    },
    {
      "name": "No conceptual digressions",
      "description": "The document does NOT include a section explaining background concepts (e.g., 'How tokenisation works', 'Understanding stdin/stdout') that is not directly needed to complete the steps",
      "max_score": 8
    }
  ]
}

evals

SKILL.md

tile.json