CtrlK
BlogDocsLog inGet started
Tessl Logo

oh-my-ai/humaniser

Removes common AI-writing tells and rewrites prose to sound natural and human, using an interactive intake, voice calibration, and a second-pass audit.

100

1.00x
Quality

100%

Does it follow best practices?

Impact

100%

1.00x

Average score across 3 eval scenarios

SecuritybySnyk

Advisory

Suggest reviewing before use

Overview
Quality
Evals
Security
Files

criteria.jsonevals/scenario-1/

{
  "context": "Tests whether the agent correctly applies the hunt list to eliminate AI vocabulary clusters, significance inflation, vague attributions, chatbot residue, and signposting patterns, while preserving factual content and delivering only the rewritten text without an editorial log.",
  "type": "weighted_checklist",
  "checklist": [
    {
      "name": "No AI vocabulary clusters",
      "description": "The output does NOT contain any of: 'transformative', 'showcasing', 'underscores', 'fostering', 'synergies', 'empowers', 'leveraging', 'unwavering'",
      "max_score": 12
    },
    {
      "name": "No significance inflation",
      "description": "The output does NOT contain any of: 'transformative', 'pivotal', 'cutting-edge', 'unparalleled', 'best-in-class', 'rapidly evolving', 'forefront'",
      "max_score": 10
    },
    {
      "name": "No vague attribution",
      "description": "The output does NOT contain 'Experts agree' or 'experts agree' (or equivalent unattributed authority claim)",
      "max_score": 10
    },
    {
      "name": "No chatbot residue",
      "description": "The output does NOT contain any of: 'Great news', 'incredibly excited', 'I hope this helps', 'Let me know if'",
      "max_score": 10
    },
    {
      "name": "No signposting openers",
      "description": "The output does NOT begin with a meta-commentary opener such as 'Here's what you need to know', 'Let's dive in', or similar signposting phrase",
      "max_score": 8
    },
    {
      "name": "No assistant preamble",
      "description": "The output file (rewritten_announcement.txt) begins directly with the rewritten content — no introductory phrase from the agent such as 'Here is the rewritten version' or 'I've rewritten the announcement'",
      "max_score": 10
    },
    {
      "name": "Preserved price figure",
      "description": "The output contains exactly '$299/seat/month' (the pricing figure is unchanged)",
      "max_score": 8
    },
    {
      "name": "Preserved product name",
      "description": "The output contains 'DataBridge Pro 3.0' (the product name and version are unchanged)",
      "max_score": 6
    },
    {
      "name": "No fake specificity added",
      "description": "The output does NOT introduce any invented statistics, study citations, or specific numbers that were not in the original text",
      "max_score": 10
    },
    {
      "name": "No new clichés",
      "description": "The output does NOT contain common replacement clichés such as 'at the end of the day', 'north star', 'game changer', 'move the needle', or 'low-hanging fruit'",
      "max_score": 8
    },
    {
      "name": "No change log in output",
      "description": "The output file contains only the rewritten text — no bullet list of changes, no edit summary, no section labelled 'Changes made' or similar",
      "max_score": 8
    }
  ]
}

evals

scenario-1

criteria.json

task.md

SKILL.md

tile.json