theclaymethod__unslop/evals/evals.json

{
  "skill_name": "unslop",
  "evals": [
    {
      "id": 1,
      "prompt": "Humanize this LinkedIn post:\n\nHere's the thing: building a startup is hard. Not because the market is competitive. Because execution is relentless. Let that sink in.\n\nIn today's fast-paced business landscape, the most successful founders are those who lean into discomfort and navigate uncertainty with clarity. They leverage their core competencies while fostering a culture of innovation.\n\nThe key takeaway? It's not about having the best idea. It's about having the best team. Full stop.\n\n#startups #leadership #entrepreneurship",
      "expected_output": "Cleaned text with zero throat-clearing openers, zero emphasis crutches, zero em-dashes, no business jargon (leverage, foster, navigate, landscape, lean into), hashtags preserved. Should be 40-60% shorter. Voice: crisp preset by default.",
      "files": [],
      "assertions": [
        {
          "name": "no_throat_clearing",
          "type": "banned_phrase_absent",
          "check": "Output contains zero instances of: 'here's the thing', 'let that sink in', 'full stop', 'the key takeaway'"
        },
        {
          "name": "no_jargon",
          "type": "banned_phrase_absent",
          "check": "Output contains zero instances of: 'leverage', 'navigate', 'landscape', 'lean into', 'foster', 'core competencies'"
        },
        {
          "name": "hashtags_preserved",
          "type": "content_preserved",
          "check": "Output preserves #startups #leadership #entrepreneurship"
        },
        {
          "name": "word_reduction",
          "type": "quantitative",
          "check": "Output word count is 40-70% of input word count"
        }
      ]
    },
    {
      "id": 2,
      "prompt": "De-slop this technical article paragraph. It has important data that must be preserved exactly:\n\nIt's worth noting that in Q3 2024, Acme Corp (NYSE: ACME) reported revenue of $47.3M, which represents a truly impressive 23% year-over-year increase. This groundbreaking result underscores the company's commitment to innovation, showcasing the effectiveness of their new API product launched in March 2024. CEO Jane Chen stated: \"We're just getting started.\" The company now serves 2,500 enterprise customers across 47 countries, positioning itself at the forefront of the cloud infrastructure landscape.",
      "expected_output": "All facts preserved exactly ($47.3M, 23%, Q3 2024, NYSE: ACME, Acme Corp, Jane Chen, March 2024, 2,500 enterprise customers, 47 countries, the direct quote). No filler (it's worth noting), no inflation (groundbreaking, at the forefront), no superficial -ing (underscoring, showcasing, positioning). Should be noticeably shorter.",
      "files": [],
      "assertions": [
        {
          "name": "revenue_preserved",
          "type": "content_preserved",
          "check": "Output contains '$47.3M'"
        },
        {
          "name": "percentage_preserved",
          "type": "content_preserved",
          "check": "Output contains '23%'"
        },
        {
          "name": "date_preserved",
          "type": "content_preserved",
          "check": "Output contains 'Q3 2024'"
        },
        {
          "name": "ceo_quote_preserved",
          "type": "content_preserved",
          "check": "Output contains the quote 'We're just getting started' attributed to Jane Chen"
        },
        {
          "name": "customer_count_preserved",
          "type": "content_preserved",
          "check": "Output contains '2,500' and '47 countries'"
        },
        {
          "name": "no_filler_or_inflation",
          "type": "banned_phrase_absent",
          "check": "Output contains zero instances of: 'it's worth noting', 'groundbreaking', 'underscores', 'showcasing', 'at the forefront', 'landscape', 'positioning'"
        },
        {
          "name": "no_superficial_ing",
          "type": "banned_phrase_absent",
          "check": "No trailing participial clauses (', underscoring...', ', showcasing...', ', positioning...')"
        }
      ]
    },
    {
      "id": 3,
      "prompt": "Make this email sound human. Use the warm preset:\n\nSubject: Following Up on Our Discussion\n\nDear Michael,\n\nI hope this email finds you well! I wanted to circle back on our discussion from Tuesday's meeting regarding the Q1 budget allocation of $2.4M for the engineering department.\n\nHere's the thing: I think it's important to note that our team has been navigating some challenges with the current timeline. The deadline of March 15, 2025 is fast approaching, and I wanted to touch base about the three deliverables we discussed:\n\n1. The API migration to v3.0\n2. The security audit (scheduled for February 28)\n3. The hiring plan for 5 senior engineers\n\nI'd be happy to schedule a follow-up meeting at your convenience. Please let me know if you need anything else!\n\nBest regards,\nSarah",
      "expected_output": "Warm, conversational tone. All facts preserved (Michael, $2.4M, Q1, March 15 2025, v3.0, February 28, 5 senior engineers, Sarah, Tuesday). No chatbot artifacts (I hope this email finds you well, I'd be happy to, let me know if you need anything else). No jargon (circle back, touch base, navigating challenges). The three deliverables must all survive.",
      "files": [],
      "assertions": [
        {
          "name": "budget_preserved",
          "type": "content_preserved",
          "check": "Output contains '$2.4M'"
        },
        {
          "name": "deadline_preserved",
          "type": "content_preserved",
          "check": "Output contains 'March 15' (with or without year)"
        },
        {
          "name": "all_deliverables_present",
          "type": "content_preserved",
          "check": "Output mentions all three: API migration/v3.0, security audit/February 28, hiring/5 senior engineers"
        },
        {
          "name": "names_preserved",
          "type": "content_preserved",
          "check": "Output contains 'Michael' and 'Sarah'"
        },
        {
          "name": "no_chatbot_artifacts",
          "type": "banned_phrase_absent",
          "check": "Output contains zero instances of: 'I hope this email finds you well', 'I'd be happy to', 'let me know if you need anything else', 'at your convenience'"
        },
        {
          "name": "no_jargon",
          "type": "banned_phrase_absent",
          "check": "Output contains zero instances of: 'circle back', 'touch base', 'navigating'"
        },
        {
          "name": "warm_tone",
          "type": "qualitative",
          "check": "Uses contractions, addresses recipient naturally, doesn't sound robotic or overly formal"
        }
      ]
    },
    {
      "id": 4,
      "prompt": "Fix this so it doesn't sound like ChatGPT wrote it:\n\nThe Apollo program stands as a testament to human ingenuity, leaving an indelible mark on the rich tapestry of space exploration. Not only did it showcase America's robust capabilities, but it also underscored the importance of fostering international collaboration. Nestled in the heart of Houston, NASA's Johnson Space Center boasts a world-class facility that serves as a beacon of scientific achievement. The program's legacy continues to shape the landscape of modern aerospace, paving the way for groundbreaking missions to Mars and beyond. Experts argue that this pivotal moment in history sends a clear message: the future looks bright for space exploration. Only time will tell what exciting discoveries lie ahead.",
      "expected_output": "Replace abstract praise with concrete facts (12 astronauts on the moon, 1969-1972, actual Houston address or description). Zero significance inflation, zero promotional language, zero generic conclusions. No copula avoidance (stands as, serves as). No superficial -ing analyses. No vague attributions.",
      "files": [],
      "assertions": [
        {
          "name": "no_significance_inflation",
          "type": "banned_phrase_absent",
          "check": "Zero instances of: 'stands as a testament', 'indelible mark', 'rich tapestry', 'beacon of', 'pivotal moment', 'groundbreaking', 'sends a clear message'"
        },
        {
          "name": "no_promotional",
          "type": "banned_phrase_absent",
          "check": "Zero instances of: 'nestled', 'boasts', 'world-class', 'in the heart of'"
        },
        {
          "name": "no_generic_conclusion",
          "type": "banned_phrase_absent",
          "check": "Zero instances of: 'the future looks bright', 'only time will tell', 'exciting discoveries lie ahead', 'continues to shape'"
        },
        {
          "name": "no_copula_avoidance",
          "type": "banned_phrase_absent",
          "check": "Zero instances of: 'stands as', 'serves as'"
        },
        {
          "name": "has_specific_facts",
          "type": "qualitative",
          "check": "Contains at least one specific fact about Apollo (dates, number of missions, astronaut count, etc.) instead of abstract praise"
        }
      ]
    }
  ]
}