cd /tmp/workspace && python3 -c " from pptx import Presentation from pptx.util import Emu prs = Presentation('/tmp/workspace/attachments/d1366f40-4803-4e3c-b140-30666c0eafd2/Precocious_Puberty_Light_v3 (1).pptx') for i, slide in enumerate(prs.slides, 1): print(f'--- slide {i} (layout: {slide.slide_layout.name}) ---') for shape in slide.shapes: if shape.has_text_frame and shape.text_frame.text.strip(): print(f' [{shape.name}]: {shape.text_frame.text[:200]}') "

mkdir -p /tmp/workspace/precocious_puberty && python /tmp/skills/pptx/scripts/office/unpack.py '/tmp/workspace/attachments/d1366f40-4803-4e3c-b140-30666c0eafd2/Precocious_Puberty_Light_v3 (1).pptx' /tmp/workspace/precocious_puberty/unpacked/ 2>&1 | tail -20

ls /tmp/workspace/precocious_puberty/unpacked/ppt/theme/ 2>/dev/null; ls /tmp/workspace/precocious_puberty/unpacked/ppt/ 2>/dev/null

python3 -c " import os, glob # Read theme XML with open('/tmp/workspace/precocious_puberty/unpacked/ppt/theme/theme1.xml') as f: content = f.read() print('THEME (first 3000 chars):') print(content[:3000]) "

python3 -c " with open('/tmp/workspace/precocious_puberty/unpacked/ppt/slides/slide1.xml') as f: content = f.read() print(content[:4000]) "

python3 -c " import os, glob, re slides_dir = '/tmp/workspace/precocious_puberty/unpacked/ppt/slides/' slides = sorted(glob.glob(slides_dir + 'slide*.xml')) print('Number of slides:', len(slides)) # Collect all unique colors across all slides all_colors = set() for sf in slides: with open(sf) as f: content = f.read() colors = re.findall(r'srgbClr val=\"([0-9A-Fa-f]{6})\"', content) all_colors.update(colors) print('Unique colors found:') for c in sorted(all_colors): print(' ', c) "

python3 << 'PYEOF' import os, glob, re slides_dir = '/tmp/workspace/precocious_puberty/unpacked/ppt/slides/' # Color mapping: original -> B&W equivalent # Dark colors -> pure black or dark greys # Light colors -> whites/light greys # Accent blues/greens -> greys color_map = { # Backgrounds 'F5F5F0': 'FFFFFF', # near-white bg -> white 'E8EAF6': 'F0F0F0', # light blue bg -> light grey 'FAF8F4': 'FFFFFF', # warm white -> white 'F1F8E9': 'EEEEEE', # light green -> light grey 'E8F5E9': 'EEEEEE', # lighter green -> light grey 'A5D6A7': 'CCCCCC', # medium green -> medium grey # Dark backgrounds / section dividers '5C6BC0': '222222', # indigo blue -> near-black '3D4EAA': '111111', # dark blue -> black '4A4A6A': '333333', # dark purple-grey -> dark grey '333344': '222222', # very dark -> near-black '37474F': '2A2A2A', # dark grey-blue -> dark grey '1C1C1C': '111111', # already near-black # Accent/highlight colors '2E7D32': '333333', # dark green -> dark grey '66BB6A': '888888', # medium green -> mid grey 'E65C00': '444444', # orange -> dark grey # Pure white stays white 'FFFFFF': 'FFFFFF', } slides = sorted(glob.glob(slides_dir + 'slide*.xml')) for sf in slides: with open(sf) as f: content = f.read() original = content for orig, bw in color_map.items(): content = content.replace(f'srgbClr val="{orig}"', f'srgbClr val="{bw}"') if content != original: with open(sf, 'w') as f: f.write(content) print('Color replacement done across', len(slides), 'slides') # Also update slide masters and layouts for xml_path in glob.glob('/tmp/workspace/precocious_puberty/unpacked/ppt/slideMasters/*.xml') + \ glob.glob('/tmp/workspace/precocious_puberty/unpacked/ppt/slideLayouts/*.xml'): with open(xml_path) as f: content = f.read() original = content for orig, bw in color_map.items(): content = content.replace(f'srgbClr val="{orig}"', f'srgbClr val="{bw}"') if content != original: with open(xml_path, 'w') as f: f.write(content) print('Color replacement done in masters/layouts too') PYEOF

python3 -c " with open('/tmp/workspace/precocious_puberty/unpacked/ppt/slides/slide38.xml') as f: content = f.read() print(content[:6000]) "

python3 -c " with open('/tmp/workspace/precocious_puberty/unpacked/ppt/slides/slide38.xml') as f: content = f.read() print(content[6000:]) "

python3 << 'PYEOF' import os # Slide 38: Discussion - Prevalence Findings (long paragraph) slide38_xml = '''<?xml version="1.0" encoding="utf-8"?> <p:sld xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main" xmlns:p="http://schemas.openxmlformats.org/presentationml/2006/main" xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships"> <p:cSld> <p:spTree> <p:nvGrpSpPr> <p:cNvPr id="1" name=""/> <p:cNvGrpSpPr/> <p:nvPr/> </p:nvGrpSpPr> <p:grpSpPr/> <p:sp> <p:nvSpPr> <p:cNvPr id="2" name="Rectangle 1"/> <p:cNvSpPr/> <p:nvPr/> </p:nvSpPr> <p:spPr> <a:xfrm> <a:off x="0" y="0"/> <a:ext cx="12191695" cy="6858000"/> </a:xfrm> <a:prstGeom prst="rect"><a:avLst/></a:prstGeom> <a:solidFill><a:srgbClr val="FFFFFF"/></a:solidFill> <a:ln><a:noFill/></a:ln> </p:spPr> <p:txBody> <a:bodyPr rtlCol="0" anchor="ctr"/> <a:lstStyle/> <a:p><a:pPr algn="ctr"/></a:p> </p:txBody> </p:sp> <p:sp> <p:nvSpPr> <p:cNvPr id="3" name="Rectangle 2"/> <p:cNvSpPr/> <p:nvPr/> </p:nvSpPr> <p:spPr> <a:xfrm> <a:off x="0" y="0"/> <a:ext cx="12191695" cy="1188720"/> </a:xfrm> <a:prstGeom prst="rect"><a:avLst/></a:prstGeom> <a:solidFill><a:srgbClr val="F0F0F0"/></a:solidFill> <a:ln><a:noFill/></a:ln> </p:spPr> <p:txBody> <a:bodyPr rtlCol="0" anchor="ctr"/> <a:lstStyle/> <a:p><a:pPr algn="ctr"/></a:p> </p:txBody> </p:sp> <p:sp> <p:nvSpPr> <p:cNvPr id="4" name="Rectangle 3"/> <p:cNvSpPr/> <p:nvPr/> </p:nvSpPr> <p:spPr> <a:xfrm> <a:off x="0" y="1188720"/> <a:ext cx="12191695" cy="64008"/> </a:xfrm> <a:prstGeom prst="rect"><a:avLst/></a:prstGeom> <a:solidFill><a:srgbClr val="333333"/></a:solidFill> <a:ln><a:noFill/></a:ln> </p:spPr> <p:txBody> <a:bodyPr rtlCol="0" anchor="ctr"/> <a:lstStyle/> <a:p><a:pPr algn="ctr"/></a:p> </p:txBody> </p:sp> <p:sp> <p:nvSpPr> <p:cNvPr id="5" name="TextBox 4"/> <p:cNvSpPr txBox="1"/> <p:nvPr/> </p:nvSpPr> <p:spPr> <a:xfrm> <a:off x="365760" y="73152"/> <a:ext cx="11430000" cy="777240"/> </a:xfrm> <a:prstGeom prst="rect"><a:avLst/></a:prstGeom> <a:noFill/> </p:spPr> <p:txBody> <a:bodyPr wrap="square"><a:spAutoFit/></a:bodyPr> <a:lstStyle/> <a:p> <a:pPr algn="l"/> <a:r> <a:rPr sz="3200" b="1" i="0"> <a:solidFill><a:srgbClr val="111111"/></a:solidFill> </a:rPr> <a:t>Discussion &#x2013; Prevalence Findings</a:t> </a:r> </a:p> </p:txBody> </p:sp> <p:sp> <p:nvSpPr> <p:cNvPr id="6" name="TextBox Body"/> <p:cNvSpPr txBox="1"/> <p:nvPr/> </p:nvSpPr> <p:spPr> <a:xfrm> <a:off x="365760" y="1380000"/> <a:ext cx="11430000" cy="5200000"/> </a:xfrm> <a:prstGeom prst="rect"><a:avLst/></a:prstGeom> <a:noFill/> </p:spPr> <p:txBody> <a:bodyPr wrap="square"> <a:spAutoFit/> </a:bodyPr> <a:lstStyle/> <a:p> <a:pPr algn="l" indent="0" marL="0"/> <a:r> <a:rPr sz="1600" b="0" i="0"> <a:solidFill><a:srgbClr val="111111"/></a:solidFill> </a:rPr> <a:t>The overall prevalence of precocious puberty (PP) in this study was 9.1% (95% CI: 6.5&#x2013;12.3%), based on questionnaire-assessed onset of secondary sexual characteristics before age 8 combined with confirmation by a registered medical practitioner. This figure places Malappuram district firmly within the range reported from other Indian urban and semi-urban settings. Binu et al. from Kollam, Kerala documented a PP prevalence of 10.4%, which is strikingly close to the present estimate, suggesting a consistent regional pattern across Kerala. Studies from major Indian metropolitan centres such as Delhi, Mumbai, and Kolkata have reported PP prevalence ranging from 5% to 12% among schoolgirls, a band within which our finding comfortably sits. By contrast, Western clinical series tend to report prevalence figures as low as 0.2%, but these figures are drawn from clinically referred populations rather than community or school-based samples and are therefore expected to substantially underestimate true population-level prevalence. The non-significance of the urban versus semi-urban difference in this study (urban 11.2% vs semi-urban 7.0%; chi-square p = 0.184) warrants careful interpretation. This finding may reflect genuine attenuation of the urban-rural gradient in Malappuram, a district undergoing rapid urbanisation in which even semi-urban areas are increasingly exposed to processed foods, sedentary lifestyles, and environmental endocrine disruptors. Alternatively, the study may have been underpowered to detect a modest difference of this magnitude between strata of roughly equal size. Despite the non-significant difference, the directional trend is consistent with urbanisation acting as a PP risk amplifier through multiple convergent pathways including dietary transition, reduced physical activity, and heightened environmental chemical exposure. Taken together, the prevalence findings affirm that PP is a quantifiable and clinically meaningful public health burden among school girls in Malappuram, with nearly one in ten girls affected and the majority of cases going undetected without systematic screening.</a:t> </a:r> </a:p> </p:txBody> </p:sp> </p:spTree> </p:cSld> <p:clrMapOvr> <a:masterClrMapping/> </p:clrMapOvr> </p:sld>''' with open('/tmp/workspace/precocious_puberty/unpacked/ppt/slides/slide38.xml', 'w') as f: f.write(slide38_xml) print("slide38.xml written") # Slide 39: Discussion - BMI & Family History slide39_xml = '''<?xml version="1.0" encoding="utf-8"?> <p:sld xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main" xmlns:p="http://schemas.openxmlformats.org/presentationml/2006/main" xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships"> <p:cSld> <p:spTree> <p:nvGrpSpPr> <p:cNvPr id="1" name=""/> <p:cNvGrpSpPr/> <p:nvPr/> </p:nvGrpSpPr> <p:grpSpPr/> <p:sp> <p:nvSpPr> <p:cNvPr id="2" name="Rectangle 1"/> <p:cNvSpPr/> <p:nvPr/> </p:nvSpPr> <p:spPr> <a:xfrm> <a:off x="0" y="0"/> <a:ext cx="12191695" cy="6858000"/> </a:xfrm> <a:prstGeom prst="rect"><a:avLst/></a:prstGeom> <a:solidFill><a:srgbClr val="FFFFFF"/></a:solidFill> <a:ln><a:noFill/></a:ln> </p:spPr> <p:txBody> <a:bodyPr rtlCol="0" anchor="ctr"/> <a:lstStyle/> <a:p><a:pPr algn="ctr"/></a:p> </p:txBody> </p:sp> <p:sp> <p:nvSpPr> <p:cNvPr id="3" name="Rectangle 2"/> <p:cNvSpPr/> <p:nvPr/> </p:nvSpPr> <p:spPr> <a:xfrm> <a:off x="0" y="0"/> <a:ext cx="12191695" cy="1188720"/> </a:xfrm> <a:prstGeom prst="rect"><a:avLst/></a:prstGeom> <a:solidFill><a:srgbClr val="F0F0F0"/></a:solidFill> <a:ln><a:noFill/></a:ln> </p:spPr> <p:txBody> <a:bodyPr rtlCol="0" anchor="ctr"/> <a:lstStyle/> <a:p><a:pPr algn="ctr"/></a:p> </p:txBody> </p:sp> <p:sp> <p:nvSpPr> <p:cNvPr id="4" name="Rectangle 3"/> <p:cNvSpPr/> <p:nvPr/> </p:nvSpPr> <p:spPr> <a:xfrm> <a:off x="0" y="1188720"/> <a:ext cx="12191695" cy="64008"/> </a:xfrm> <a:prstGeom prst="rect"><a:avLst/></a:prstGeom> <a:solidFill><a:srgbClr val="333333"/></a:solidFill> <a:ln><a:noFill/></a:ln> </p:spPr> <p:txBody> <a:bodyPr rtlCol="0" anchor="ctr"/> <a:lstStyle/> <a:p><a:pPr algn="ctr"/></a:p> </p:txBody> </p:sp> <p:sp> <p:nvSpPr> <p:cNvPr id="5" name="TextBox 4"/> <p:cNvSpPr txBox="1"/> <p:nvPr/> </p:nvSpPr> <p:spPr> <a:xfrm> <a:off x="365760" y="73152"/> <a:ext cx="11430000" cy="777240"/> </a:xfrm> <a:prstGeom prst="rect"><a:avLst/></a:prstGeom> <a:noFill/> </p:spPr> <p:txBody> <a:bodyPr wrap="square"><a:spAutoFit/></a:bodyPr> <a:lstStyle/> <a:p> <a:pPr algn="l"/> <a:r> <a:rPr sz="3200" b="1" i="0"> <a:solidFill><a:srgbClr val="111111"/></a:solidFill> </a:rPr> <a:t>Discussion &#x2013; BMI &amp; Family History as Independent Predictors</a:t> </a:r> </a:p> </p:txBody> </p:sp> <p:sp> <p:nvSpPr> <p:cNvPr id="6" name="TextBox Body"/> <p:cNvSpPr txBox="1"/> <p:nvPr/> </p:nvSpPr> <p:spPr> <a:xfrm> <a:off x="365760" y="1380000"/> <a:ext cx="11430000" cy="5200000"/> </a:xfrm> <a:prstGeom prst="rect"><a:avLst/></a:prstGeom> <a:noFill/> </p:spPr> <p:txBody> <a:bodyPr wrap="square"> <a:spAutoFit/> </a:bodyPr> <a:lstStyle/> <a:p> <a:pPr algn="l" indent="0" marL="0"/> <a:r> <a:rPr sz="1600" b="0" i="0"> <a:solidFill><a:srgbClr val="111111"/></a:solidFill> </a:rPr> <a:t>Among all the risk factors examined in this study, elevated BMI and a positive family history of early puberty emerged as the two strongest and most statistically robust independent predictors of precocious puberty in the logistic regression model. The mean BMI of girls in the PP group was significantly higher at 21.24 kg/m&#xB2; compared to 18.89 kg/m&#xB2; in the non-PP group (t = 5.670; p &lt; 0.001), and the association was even more striking when examined categorically: among the small but important group of overweight or obese girls (BMI &#x2265; 25.0 kg/m&#xB2;), the PP rate reached 71.4%, compared to just 5.3% among underweight girls and 10.3% among those with normal BMI (chi-square p &lt; 0.001). This biologically coherent dose-response pattern is well-supported by the mechanistic literature. Excess adipose tissue generates elevated circulating leptin, which lowers the threshold for hypothalamic GnRH pulsatility and thereby permits premature activation of the hypothalamic-pituitary-gonadal (HPG) axis. Additionally, insulin and IGF-1 signalling pathways, both upregulated in obesity, provide further stimulatory input to the HPG axis, while peripheral aromatisation of androgens to oestrogens in adipose tissue directly raises systemic oestrogen levels. These converging mechanisms collectively explain why even modest increases in body fatness can accelerate pubertal onset, a finding consistently replicated by Wang et al. (2025) in their systematic review and by Gonc and Kandemir (2022) in their review of body composition and sexual precocity. Regarding family history, girls with a positive family history of early puberty had a PP rate of 20.3%, compared to only 6.6% among those without such history, yielding a highly significant chi-square statistic (p &lt; 0.001) and an adjusted odds ratio of 3.31 (p = 0.002) in logistic regression. This finding aligns with the well-established heritability of pubertal timing, estimated at 50 to 80% in twin and family studies. Specific causal mutations in genes including MKRN3, DLK1, and KISS1R have been identified in familial CPP cases by Kentistou et al. (2024) and Cheuiche et al. (2021). The strong predictive value of family history in this population highlights it as the single most actionable screening marker available in a resource-limited school health setting, requiring no equipment or laboratory testing to identify. Together, BMI and family history account for the majority of attributable risk in this sample and represent the core targets for both individual-level clinical intervention and population-level preventive strategy.</a:t> </a:r> </a:p> </p:txBody> </p:sp> </p:spTree> </p:cSld> <p:clrMapOvr> <a:masterClrMapping/> </p:clrMapOvr> </p:sld>''' with open('/tmp/workspace/precocious_puberty/unpacked/ppt/slides/slide39.xml', 'w') as f: f.write(slide39_xml) print("slide39.xml written") # Slide 40: Discussion - Diet, Screen Time & Pesticide Exposure slide40_xml = '''<?xml version="1.0" encoding="utf-8"?> <p:sld xmlns:a="http://schemas.openxmlformats.org/drawingml/2006/main" xmlns:p="http://schemas.openxmlformats.org/presentationml/2006/main" xmlns:r="http://schemas.openxmlformats.org/officeDocument/2006/relationships"> <p:cSld> <p:spTree> <p:nvGrpSpPr> <p:cNvPr id="1" name=""/> <p:cNvGrpSpPr/> <p:nvPr/> </p:nvGrpSpPr> <p:grpSpPr/> <p:sp> <p:nvSpPr> <p:cNvPr id="2" name="Rectangle 1"/> <p:cNvSpPr/> <p:nvPr/> </p:nvSpPr> <p:spPr> <a:xfrm> <a:off x="0" y="0"/> <a:ext cx="12191695" cy="6858000"/> </a:xfrm> <a:prstGeom prst="rect"><a:avLst/></a:prstGeom> <a:solidFill><a:srgbClr val="FFFFFF"/></a:solidFill> <a:ln><a:noFill/></a:ln> </p:spPr> <p:txBody> <a:bodyPr rtlCol="0" anchor="ctr"/> <a:lstStyle/> <a:p><a:pPr algn="ctr"/></a:p> </p:txBody> </p:sp> <p:sp> <p:nvSpPr> <p:cNvPr id="3" name="Rectangle 2"/> <p:cNvSpPr/> <p:nvPr/> </p:nvSpPr> <p:spPr> <a:xfrm> <a:off x="0" y="0"/> <a:ext cx="12191695" cy="1188720"/> </a:xfrm> <a:prstGeom prst="rect"><a:avLst/></a:prstGeom> <a:solidFill><a:srgbClr val="F0F0F0"/></a:solidFill> <a:ln><a:noFill/></a:ln> </p:spPr> <p:txBody> <a:bodyPr rtlCol="0" anchor="ctr"/> <a:lstStyle/> <a:p><a:pPr algn="ctr"/></a:p> </p:txBody> </p:sp> <p:sp> <p:nvSpPr> <p:cNvPr id="4" name="Rectangle 3"/> <p:cNvSpPr/> <p:nvPr/> </p:nvSpPr> <p:spPr> <a:xfrm> <a:off x="0" y="1188720"/> <a:ext cx="12191695" cy="64008"/> </a:xfrm> <a:prstGeom prst="rect"><a:avLst/></a:prstGeom> <a:solidFill><a:srgbClr val="333333"/></a:solidFill> <a:ln><a:noFill/></a:ln> </p:spPr> <p:txBody> <a:bodyPr rtlCol="0" anchor="ctr"/> <a:lstStyle/> <a:p><a:pPr algn="ctr"/></a:p> </p:txBody> </p:sp> <p:sp> <p:nvSpPr> <p:cNvPr id="5" name="TextBox 4"/> <p:cNvSpPr txBox="1"/> <p:nvPr/> </p:nvSpPr> <p:spPr> <a:xfrm> <a:off x="365760" y="73152"/> <a:ext cx="11430000" cy="777240"/> </a:xfrm> <a:prstGeom prst="rect"><a:avLst/></a:prstGeom> <a:noFill/> </p:spPr> <p:txBody> <a:bodyPr wrap="square"><a:spAutoFit/></a:bodyPr> <a:lstStyle/> <a:p> <a:pPr algn="l"/> <a:r> <a:rPr sz="3200" b="1" i="0"> <a:solidFill><a:srgbClr val="111111"/></a:solidFill> </a:rPr> <a:t>Discussion &#x2013; Diet, Screen Time &amp; Pesticide Exposure</a:t> </a:r> </a:p> </p:txBody> </p:sp> <p:sp> <p:nvSpPr> <p:cNvPr id="6" name="TextBox Body"/> <p:cNvSpPr txBox="1"/> <p:nvPr/> </p:nvSpPr> <p:spPr> <a:xfrm> <a:off x="365760" y="1380000"/> <a:ext cx="11430000" cy="5200000"/> </a:xfrm> <a:prstGeom prst="rect"><a:avLst/></a:prstGeom> <a:noFill/> </p:spPr> <p:txBody> <a:bodyPr wrap="square"> <a:spAutoFit/> </a:bodyPr> <a:lstStyle/> <a:p> <a:pPr algn="l" indent="0" marL="0"/> <a:r> <a:rPr sz="1600" b="0" i="0"> <a:solidFill><a:srgbClr val="111111"/></a:solidFill> </a:rPr> <a:t>Three additional exposure categories &#x2014; dietary patterns, screen time, and pesticide or chemical exposure &#x2014; were examined as potential risk factors but none reached statistical significance in the bivariate analysis. Regarding dietary patterns, the high prevalence of fast food consumption exceeding three times per week (46.1%) and processed food intake (51.5%) in this study population is epidemiologically concerning even in the absence of a statistically significant association with PP in the current dataset. The most likely explanation for this null finding is that this study was powered for prevalence estimation rather than for dietary subgroup analyses, meaning it lacked the statistical resolution to detect modest dietary effects. Furthermore, dietary variables were coded as simple dichotomous indicators, a methodological approach that fails to capture the dose-response gradient, nutrient composition, or duration of exposure that would be needed to reveal biologically meaningful relationships. The literature strongly implicates phthalates and bisphenol A (BPA) present in food packaging and plastic containers as dietary endocrine disruptors capable of advancing pubertal onset, as demonstrated by Calcaterra et al. (2024) and Lopez-Rodriguez et al. (2021), but these chemical exposures require urinary metabolite assays rather than questionnaire data to quantify reliably. With respect to screen time, a large proportion of the sample &#x2014; 59.0% by parental report and 62.5% by student self-report &#x2014; exceeded the World Health Organization&#x2019;s recommended two-hour daily limit for school-age children. Despite this high prevalence, no significant association with PP was found (p = 0.860). The proposed biological pathways linking screen time to earlier puberty, specifically through obesity promotion and circadian disruption due to evening blue light exposure suppressing melatonin, are indirect and would require continuous quantitative measurement of screen exposure rather than a binary cut-off to be properly evaluated. Pesticide and chemical exposure, reported by 14.1% of participants and particularly relevant to the semi-urban Irumbuzhi site where active agricultural activity occurs, similarly showed no statistically significant association with PP (p = 1.000). This null result most likely reflects the limitations of binary self-reported exposure data that captured no information about the type, duration, frequency, or intensity of chemical contact. Organochlorine pesticides, known endocrine disruptors identified by Symeonides et al. (2024) in their umbrella review, require objective biomarker measurement through serum or urinary assays to properly evaluate their impact on pubertal timing. The collective absence of significance for these three factors should not be interpreted as absence of biological effect; rather, these findings underscore the methodological constraints of the current study design and point clearly toward the need for future investigations incorporating objective dietary biomarkers, continuous screen time monitoring, and validated chemical exposure assessments in this population.</a:t> </a:r> </a:p> </p:txBody> </p:sp> </p:spTree> </p:cSld> <p:clrMapOvr> <a:masterClrMapping/> </p:clrMapOvr> </p:sld>''' with open('/tmp/workspace/precocious_puberty/unpacked/ppt/slides/slide40.xml', 'w') as f: f.write(slide40_xml) print("slide40.xml written") PYEOF

python /tmp/skills/pptx/scripts/clean.py /tmp/workspace/precocious_puberty/unpacked/ 2>&1 && echo "CLEAN OK"

python /tmp/skills/pptx/scripts/office/pack.py /tmp/workspace/precocious_puberty/unpacked/ /tmp/workspace/precocious_puberty/Precocious_Puberty_BW.pptx --original '/tmp/workspace/attachments/d1366f40-4803-4e3c-b140-30666c0eafd2/Precocious_Puberty_Light_v3 (1).pptx' 2>&1 && echo "PACK OK"

# Quick verification: confirm discussion slides have paragraph content python3 -c " from pptx import Presentation prs = Presentation('/tmp/workspace/precocious_puberty/Precocious_Puberty_BW.pptx') for i in [38, 39, 40]: slide = prs.slides[i-1] print(f'--- Slide {i} ---') for shape in slide.shapes: if shape.has_text_frame: txt = shape.text_frame.text.strip() if txt: print(f' [{shape.name}] ({len(txt)} chars): {txt[:120]}...' if len(txt) > 120 else f' [{shape.name}]: {txt}') "