CtrlK
BlogDocsLog inGet started
Tessl Logo

tessl/review-plugin-creator

Guided workflow for creating a custom Tessl reviewer plugin, by forking the default rubric or building one from scratch. Scaffolds the plugin directory structure, authors rubrics and config.json, and validates the result with tessl review run.

95

1.49x
Quality

93%

Does it follow best practices?

Impact

100%

1.49x

Average score across 4 eval scenarios

SecuritybySnyk

Passed

No known issues

Overview
Quality
Evals
Security
Files

description.jsonskills/create-review-plugin/references/default-rubric/

{
  "$schema": "../schemas/rubric.schema.json",
  "evaluation_target": "description",
  "scale": {
    "min": 1,
    "max": 3
  },
  "reference_examples": {
    "judging_guidelines": [
      "Score only what is explicitly stated; do not infer capabilities or intentions.",
      "Penalize vague fluff, buzzwords, and over-claims.",
      "Do not reward verbosity; concise and clear beats long and padded.",
      "Evaluate trigger terms as: 'would a user naturally say this when they need this skill?'",
      "A missing 'Use when...' clause or equivalent explicit trigger guidance should cap completeness at 2.",
      "Descriptions must use third person voice ('Processes files', 'Analyzes data'). Penalize first person ('I can help') or second person ('You can use this') by reducing specificity score by 1."
    ],
    "good_overall_examples": [
      "Extract text and tables from PDF files, fill forms, merge documents. Use when working with PDF files or when the user mentions PDFs, forms, or document extraction.",
      "Analyze Excel spreadsheets, create pivot tables, generate charts. Use when analyzing Excel files, spreadsheets, tabular data, or .xlsx files.",
      "Generate descriptive commit messages by analyzing git diffs. Use when the user asks for help writing commit messages or reviewing staged changes.",
      "Presentation creation, editing, and analysis. When Claude needs to work with presentations (.pptx files) for: (1) Creating new presentations, (2) Modifying or editing content, (3) Working with layouts, (4) Adding comments or speaker notes, or any other presentation tasks.",
      "Applies Anthropic's official brand colors and typography to any sort of artifact that may benefit from having Anthropic's look-and-feel. Use it when brand colors or style guidelines, visual formatting, or company design standards apply."
    ],
    "bad_overall_examples": [
      "Helps with documents",
      "Processes data",
      "Does stuff with files"
    ]
  },
  "dimensions": [
    {
      "id": "specificity",
      "name": "Specificity of Capabilities",
      "weight": 0.2,
      "question": "Does it describe concrete actions vs vague language?",
      "scores": [
        {
          "score": 1,
          "anchor": "Vague or no actions; abstract language",
          "example": "Helps with documents"
        },
        {
          "score": 2,
          "anchor": "Names domain and some actions, but not comprehensive",
          "example": "Processes PDF files and extracts content"
        },
        {
          "score": 3,
          "anchor": "Lists multiple specific concrete actions",
          "example": "Extract text and tables from PDF files, fill forms, merge documents"
        }
      ]
    },
    {
      "id": "trigger_term_quality",
      "name": "Trigger Term Quality",
      "weight": 0.3,
      "question": "Does it include natural keywords users would actually say?",
      "scores": [
        {
          "score": 1,
          "anchor": "No natural keywords; technical jargon or overly generic",
          "example": "Handles document object model manipulation"
        },
        {
          "score": 2,
          "anchor": "Some relevant keywords but missing common variations",
          "example": "Works with PDF files"
        },
        {
          "score": 3,
          "anchor": "Good coverage of natural terms users would say",
          "example": "PDF files, PDFs, forms, document extraction, .pdf"
        }
      ]
    },
    {
      "id": "completeness",
      "name": "Completeness",
      "weight": 0.35,
      "question": "Does it explicitly answer both 'what does this do' AND 'when should Claude use it'?",
      "scores": [
        {
          "score": 1,
          "anchor": "Missing what OR when, or both very weak",
          "example": "Helps with documents"
        },
        {
          "score": 2,
          "anchor": "Has what, but when is missing or only implied",
          "example": "Extract text and tables from PDF files, fill forms, merge documents"
        },
        {
          "score": 3,
          "anchor": "Clearly answers both what AND when with explicit triggers",
          "example": "Extract text and tables from PDF files, fill forms, merge documents. Use when working with PDF files or when the user mentions PDFs, forms, or document extraction."
        }
      ]
    },
    {
      "id": "distinctiveness_conflict_risk",
      "name": "Distinctiveness / Conflict Risk",
      "weight": 0.15,
      "question": "Is it clearly distinguishable from other skills; unlikely to trigger for the wrong skill?",
      "scores": [
        {
          "score": 1,
          "anchor": "Very generic; would conflict with many skills",
          "example": "Helps with code and documents"
        },
        {
          "score": 2,
          "anchor": "Somewhat specific but could still overlap with similar skills",
          "example": "Works with document files"
        },
        {
          "score": 3,
          "anchor": "Clear niche with distinct triggers; unlikely to conflict",
          "example": "Extract text and tables from PDF files, fill forms, merge documents. Use when working with PDF files or when the user mentions PDFs."
        }
      ]
    }
  ]
}

README.md

tile.json