Here's a thinking process: 1. **Analyze User Input:** - User is acting as a job listing editor. - I receive raw scraped text from a job website. - My job: extract ONLY actual job information, return clean, readable format. - Must REMOVE: social sharing text, author bylines, "you may also like", navigation, comments, ads/promotions, cookie/subscription notices, duplicate text, category labels not part of job, stray numbers/formatting artifacts. - Must KEEP and STRUCTURE: job title, company/organization, location, job description/responsibilities, requirements/qualifications, how to apply, salary (if mentioned), deadline/closing date, job ID/reference number, contract type/level. - Return ONLY the cleaned job text. No commentary, no headers like "Here is the cleaned version", no markdown. Just the text. Wait, the actual prompt I received has TWO parts: - First part: "You are a job listing editor... Return ONLY the cleaned job text." - Second part: "You receive a plain-text job description. Your ONLY job is to format it as clean semantic HTML." with specific HTML rules. - Then it gives me the job description text to process. I need to figure out which set of instructions to follow. The very first block says: "You are a job listing editor... Return ONLY the cleaned job text. Do not add any commentary, headers like "Here is the cleaned version", or markdown formatting. Just the clean job listing text." Then there's a user message with the job description text, but it's formatted with inline CSS/spans etc. Then there's another set of instructions: "You receive a plain-text job description. Your ONLY job is to format it as clean semantic HTML." with HTML tags rules. Let me re-read carefully. The user message starts with "You receive a plain-text job description. Your ONLY job is to format it as clean semantic HTML." Then gives rules for HTML formatting. Then provides the job description text. At the very end: "Return ONLY the HTML. No markdown fences, no explanation, no preamble." So the actual task is: take the provided job description text, clean it according to the rules (remove unwanted stuff, structure with HTML tags), and output ONLY the HTML. But wait, the very first block of the prompt says: "You are a job listing editor. You receive raw scraped text from a job website. Your job is to extract ONLY the actual job information and return it in a clean, readable format." Then lists what to remove/keep. Then says "Return ONLY the cleaned job text. Do not add any commentary... Just the clean job listing text." Then the second block starts: "You receive a plain-text job description. Your ONLY job is to format it as clean semantic HTML." with HTML tag rules. And ends with "Return ONLY the HTML. No markdown fences, no explanation, no preamble." I think the user wants me to follow the HTML formatting rules, because the final instruction says "Return ONLY the HTML." And the provided text is the job description to format. The first block might be context/setup, but the actual task is the HTML formatting one, as it has specific tag rules and ends with the strict output requirement. Let me check the job description text provided. It has: - `

Lets Write Africa's Story Together!

` - `

Old Mutual is a firm believer in the African opportunity and our diverse talent reflects this.

` - Some empty paragraphs - `Job Description` - `Performs simple clerical duties...` - Responsibilities section with h1/h2/h2 headings and p tags for duties - Skills section with b tags and h3 empty things - Competencies list with h3 empty - Education section with "Bachelors Degree (B)" - Closing Date: "19 August 2026 , 23:59" - "The Old Mutual Story!" with emphasis Rules for HTML formatting: - Use `

` for introductory paragraphs and general text - Use `

` for section headers (e.g. Responsibilities, Qualifications, Requirements, Education, Experience, Skills, How to Apply, Key Duties, About the Role) - Use `

• ` for list items (duties, requirements, qualifications, skills) - No CSS classes, no inline styles, no `
  `s, no `
  ` tags - Preserve ALL original content word-for-word — do not add, remove, summarise, or rephrase - If text has no clear sections, just wrap paragraphs in `

  ` tags - Return ONLY the HTML. No markdown fences, no explanation, no preamble. Important: "Preserve ALL original content word-for-word — do not add, remove, summarise, or rephrase". This is tricky because I need to restructure into HTML tags but keep all text exactly as is. The original text has inline CSS and spans which I must remove, but the text content inside must be preserved exactly. Let me parse the original text and map it to HTML structure, keeping all words. Original text blocks (cleaning out empty paragraphs and formatting artifacts): 1. `

  Lets Write Africa's Story Together!

  ` -> Keep "Lets Write Africa's Story Together!" as title? The instruction says keep job title. I'll put it as an

  or

  ? The rules say use

  for section headers, and

  for introductory paragraphs. The title "Lets Write Africa's Story Together!" seems like a job title or company tagline. I'll make it an

  maybe, or keep as

  . But rules say use

  for section headers like Responsibilities, Qualifications, etc. The title might be best as

  or

  . I need to decide. Since it's not one of the listed section headers, I'll probably wrap it in

  or maybe

  if it's the title. Let's see. 2. `

  Old Mutual is a firm believer in the African opportunity and our diverse talent reflects this.

  ` -> Company description. Keep as

  . 3. Some empty `

  ` - I'll ignore empty paragraphs or just not output them? Rules say preserve ALL original content word-for-word. But empty paragraphs have no words. I think I should omit truly empty ones, but to be safe, maybe just not include them since they contain no content. The rule "Preserve ALL original content word-for-word" means don't remove or change the actual text words, but empty tags can be dropped. 4. `Job Description` -> This is a heading. Rules say use

  for section headers. "Job Description" is a section header. I'll use

  Job Description

  . 5. `Performs simple clerical duties by applying office procedures including answering calls, managing basic financial transactions and recording, typing or word processing, filing, and other ad-hoc activities to increase efficiency.` -> This is a paragraph. Keep as

  ...text...

  . Then the responsibilities section: `

  Responsibilities

  ` -> Rules say use

  for section headers. But the original has

  . I must preserve content word-for-word, but I can change the tag? The rule says "Use

  for section headers (e.g. Responsibilities, Qualifications, ...)". It doesn't say I must use exactly the same tag; it says use

  for section headers. Since the original

  is a section header, I should convert it to

  to follow the rules. But "Preserve ALL original content word-for-word" might conflict. I think the intent is keep the text content exactly, but use the specified tags. I'll use

  Responsibilities

  . Then `

  ` empty - skip. Then `

  Administration

  ` -> use

  Administration

  ? Wait, the rules list section headers as examples: "Responsibilities, Qualifications, Requirements, Education, Experience, Skills, How to Apply, Key Duties, About the Role". Administration is a subsection under Responsibilities. In the original, it's `

  Administration

  `. Should I nest it? Rules say use

  • for list items. The responsibilities are currently in

    tags under each h2. I need to structure them. The rule: "Use

    • for list items (duties, requirements, qualifications, skills)". So I should convert the duty paragraphs into
      • items. But "Preserve ALL original content word-for-word" - I need to keep the text of each duty exactly, but wrap them in
      • . Let look at the responsibilities structure: ```

        Administration

        Conduct data entry into company systems, and file information using established procedures to ensure information is accurate and accessible.

        Document Management

        File records or documents as instructed.

        Document Preparation

        Input content into standard templates. Also take responsibility for collating materials and photocopying them.

        Work Scheduling and Allocation

        Work according to an assigned schedule.

        Operations Management

        Carry out operational tasks by following established processes. Involves working within given settings or parameters.

        Internal Client Relationship Management

        Exchange information with internal clients by having courteous interactions with them.

        Operational Compliance

        Follow a straightforward mandatory procedure to ensure that work is carried out to the required quality standards and/or to ensure adherence to external regulatory codes and internal codes of conduct.

        Data Collection and Analysis

        Perform basic data entry tasks.

        Personal Capability Building

        Develop and maintain excellent skills by participating in assessment and development planning activities as well as formal and informal training and coaching.

        ``` Then `
        ` etc. Then `Skills` section: ```
        ...

        Skills

        ...
        ...
        ...

        Competencies

        ... Collaborates
        Communicates Effectively
        Demonstrates Self-Awareness
        Drives Results
        Ensures Accountability
        Manages Complexity
        Optimizes Work Processes
        Organizational Savvy
        ... ``` Then `Education`: ```
        ...

        Education

        ... Bachelors Degree (B)
        ... ``` Then `Closing Date`: ```

        Closing Date

        ... 19 August 2026 , 23:59
        ... ``` Then `The Old Mutual Story!` at the end. Now, I need to output clean semantic HTML using only

        ,

        ,

        ,
        • . No CSS classes, no inline styles, no
          , no
          . Strategy: - Parse all text content, ignoring empty paragraphs and formatting tags. - Structure into: - Title/company intro as

          or

          - Job Description intro paragraph - Section: Responsibilities (using

          Responsibilities

          , then

          • for each duty paragraph) - Section: Skills/Competencies (maybe combine or separate; original has "Skills" and "Competencies" with lists) - Section: Education - Section: Closing Date - Any other info But I must preserve ALL original content word-for-word. So I need to include every word exactly as written, including "Lets Write Africa's Story Together!", "Old Mutual is a firm believer...", the duty texts, the competency names, "Bachelors Degree (B)", "19 August 2026 , 23:59", "The Old Mutual Story!". Let me list all distinct text segments in order, stripping the HTML formatting tags but keeping text: 1. "Lets Write Africa's Story Together!" (from first span) 2. "Old Mutual is a firm believer in the African opportunity and our diverse talent reflects this." 3. (empty paragraphs - skip) 4. "Job Description" (from Job Description) 5. "Performs simple clerical duties by applying office procedures including answering calls, managing basic financial transactions and recording, typing or word processing, filing, and other ad-hoc activities to increase efficiency." 6. (empty paragraphs - skip) 7. "Responsibilities" (but original

            , I'll use

            per rules) 8. "Administration" 9. "Conduct data entry into company systems, and file information using established procedures to ensure information is accurate and accessible." 10. (empty paragraph - skip) 11. "Document Management" 12. "File records or documents as instructed." 13. (empty paragraph - skip) 14. "Document Preparation" 15. "Input content into standard templates. Also take responsibility for collating materials and photocopying them." 16. (empty paragraph - skip) 17. "Work Scheduling and Allocation" 18. "Work according to an assigned schedule." 19. (empty paragraph - skip) 20. "Operations Management" 21. "Carry out operational tasks by following established processes. Involves working within given settings or parameters." 22. (empty paragraph - skip) 23. "Internal Client Relationship Management" 24. "Exchange information with internal clients by having courteous interactions with them." 25. (empty paragraph - skip) 26. "Operational Compliance" 27. "Follow a straightforward mandatory procedure to ensure that work is carried out to the required quality standards and/or to ensure adherence to external regulatory codes and internal codes of conduct." 28. (empty paragraph - skip) 29. "Data Collection and Analysis" 30. "Perform basic data entry tasks." 31. (empty paragraph - skip) 32. "Personal Capability Building" 33. "Develop and maintain excellent skills by participating in assessment and development planning activities as well as formal and informal training and coaching." 34. (empty paragraph - skip) 35. (some stray

            with "na" and non-breaking space? Actually: `na

             

            ` - there's "na" text. I need to preserve it? It says "na

             

            " - the "na" might be part of the text. I'll include "na" as a paragraph or part of text. Let me re-read: `na

             

            ` - probably "na" is stray text, maybe "Not applicable" or just "na". I'll keep it as is, word-for-word. I'll put it in a

            .) 36. Then empty paragraphs. 37. `Skills` -> "Skills" 38. Empty paragraphs under Skills. 39. `Competencies` -> "Competencies" 40. Then competency list items: "Collaborates", "Communicates Effectively", "Demonstrates Self-Awareness", "Drives Results", "Ensures Accountability", "Manages Complexity", "Optimizes Work Processes", "Organizational Savvy". These are under h3 tags that are empty. I'll make them

            items. 41. Then empty paragraphs. 42. `Education` -> "Education" 43. "Bachelors Degree (B)" 44. Empty paragraphs. 45. `Closing Date` -> "Closing Date" 46. "19 August 2026 ,

  Translated by AI — may not be perfect

  Show original