Integrate document analysis framework with all content generators#1238
Merged
Integrate document analysis framework with all content generators#1238
Conversation
Contributor
🏷️ Automatic Labeling SummaryThis PR has been automatically labeled based on the files changed and PR metadata. Applied Labels: size-xs Label Categories
For more information, see |
Contributor
🔍 Lighthouse Performance Audit
📥 Download full Lighthouse report Budget Compliance: Performance budgets enforced via |
- Extended DeepAnalysisOptions with optional frameworkAnalysis parameter - Added analyzeDocumentsForContent() bridge function in shared.ts - Added PESTLE, stakeholder impact, risk, and implementation section renderers when framework analysis is available - Updated 6 content generators to pass framework analysis through - Added 4 new localization keys across all 14 languages - Fixed null safety in calculateCoalitionRiskIndex for partyPerformance - All 3711 existing tests pass Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>
github-actions
Bot
added
refactor
Contributor
🔍 Lighthouse Performance Audit
📥 Download full Lighthouse report Budget Compliance: Performance budgets enforced via |
- Updated analyzeDocumentsForContent() to return both framework and perspective analyses (6 lenses: government, opposition, citizen, economic, international, media) - Extended DeepAnalysisOptions with perspectiveAnalysis field - Added perspective insights rendering to generateDeepAnalysisSection() - Fixed null safety for partyPerformance in government and opposition lenses - Updated all 6 content generators to pass both analysis results through - Added 19 comprehensive integration tests covering: - Bridge function returns both analysis types - PESTLE, stakeholder, risk, and implementation section rendering - 6-lens perspective insights rendering - Multi-language label support (en, sv, de, ja, ar) - HTML escaping / XSS prevention - Graceful degradation without framework analysis - Confidence scoring and iteration metadata - All 3730 tests pass (3711 existing + 19 new) Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>
Contributor
🔍 Lighthouse Performance Audit
📥 Download full Lighthouse report Budget Compliance: Performance budgets enforced via |
- Extracted 6 named constants (MAX_PESTLE_ITEMS, MAX_STAKEHOLDER_IMPACTS, MAX_RISK_ITEMS, MAX_PERSPECTIVE_INSIGHTS, MAX_IMPLEMENTATION_OBSTACLES, MAX_AGENCIES_DISPLAYED) for improved code maintainability - Fixed test description to match actual validation (≥3 groups) - Fixed CIAContext fixture to satisfy strict type checking - All 3730 tests pass, TypeScript compiles cleanly Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>
Contributor
🔍 Lighthouse Performance Audit
📥 Download full Lighthouse report Budget Compliance: Performance budgets enforced via |
Copilot
AI
changed the title
[WIP] Add AI-powered comprehensive document analysis framework
Integrate document analysis framework with all content generators
Mar 16, 2026
Contributor
There was a problem hiding this comment.
Pull request overview
This PR connects the existing AI document analysis framework and 6-lens perspective analysis framework to all article content generators, so deep analysis sections can be enriched with structured PESTLE, stakeholder impact, risk, implementation, and perspective insights.
Changes:
- Added a shared bridge (
analyzeDocumentsForContent()) and extended deep-analysis rendering to accept optional framework + perspective analysis inputs. - Updated all relevant content generators to call the bridge and pass analyses through to
generateDeepAnalysisSection(). - Expanded localization labels and added integration tests; also fixed null-safety around
cia.partyPerformance.
Reviewed changes
Copilot reviewed 15 out of 15 changed files in this pull request and generated 13 comments.
Show a summary per file
| File | Description |
|---|---|
| tests/analysis-integration.test.ts | New integration coverage for bridge + enriched deep analysis rendering and localization. |
| scripts/types/content.ts | Extends ContentLabelSet with new deep-analysis framework label keys. |
| scripts/data-transformers/risk-analysis.ts | Null-safe guard for partyPerformance in coalition risk calculation. |
| scripts/data-transformers/content-generators/week-ahead.ts | Uses bridge + passes framework/perspective analyses into deep analysis section. |
| scripts/data-transformers/content-generators/shared.ts | Introduces bridge + extends generateDeepAnalysisSection() to render framework and 6-lens insights. |
| scripts/data-transformers/content-generators/propositions.ts | Uses bridge + passes analyses into deep analysis rendering. |
| scripts/data-transformers/content-generators/motions.ts | Uses bridge + passes analyses into deep analysis rendering. |
| scripts/data-transformers/content-generators/interpellations.ts | Uses bridge + passes analyses into deep analysis rendering. |
| scripts/data-transformers/content-generators/index.ts | Re-exports analyzeDocumentsForContent() from the generator barrel. |
| scripts/data-transformers/content-generators/generic.ts | Uses bridge + passes analyses into deep analysis rendering. |
| scripts/data-transformers/content-generators/committee.ts | Uses bridge + passes analyses into deep analysis rendering. |
| scripts/data-transformers/constants/content-labels-part1.ts | Adds new framework label keys for part1 languages. |
| scripts/data-transformers/constants/content-labels-part2.ts | Adds new framework label keys for part2 languages. |
| scripts/analysis-framework/lenses/opposition.ts | Null-safe guard for cia.partyPerformance usage. |
| scripts/analysis-framework/lenses/government.ts | Null-safe guard for cia.partyPerformance usage. |
Comment on lines
266
to
275
| const { frameworkAnalysis, perspectiveAnalysis } = analyzeDocumentsForContent(allWeekDocs, lang, weekCia); | ||
| content += generateDeepAnalysisSection({ | ||
| documents: allWeekDocs, | ||
| lang, | ||
| cia: weekCia, | ||
| articleType: 'week-ahead', | ||
| frameworkAnalysis, | ||
| perspectiveAnalysis, | ||
| }); | ||
|
|
Comment on lines
+937
to
+941
| const top = [...byType.values()].slice(0, MAX_RISK_ITEMS); | ||
| const rows = top.map(r => { | ||
| const icon = r.severity === 'high' ? '🔴' : r.severity === 'medium' ? '🟡' : '🟢'; | ||
| return ` <li>${icon} <strong>${escapeHtml(r.type)}</strong> (${escapeHtml(r.severity)}): ${escapeHtml(r.description)}</li>`; | ||
| }); |
Comment on lines
+892
to
+905
| * Render a summary of stakeholder impacts across all analysed documents. | ||
| * Shows up to 7 stakeholder groups with their impact direction and burden. | ||
| */ | ||
| function renderStakeholderImpactSummary(analyses: DocumentAnalysis[]): string { | ||
| // Collect all stakeholder impacts, deduplicated by stakeholder name | ||
| const impactMap = new Map<string, StakeholderImpact>(); | ||
| for (const a of analyses) { | ||
| for (const impact of a.stakeholderImpacts) { | ||
| // Keep the first (or higher-magnitude) impact per stakeholder | ||
| const existing = impactMap.get(impact.stakeholder); | ||
| if (!existing) { | ||
| impactMap.set(impact.stakeholder, impact); | ||
| } | ||
| } |
Comment on lines
+971
to
+982
| const parts: string[] = []; | ||
| const fIcon = highestFeasibility === 'high' ? '🟢' : highestFeasibility === 'medium' ? '🟡' : '🔴'; | ||
| parts.push(` <p>${fIcon} <strong>Feasibility:</strong> ${escapeHtml(highestFeasibility)}. ${escapeHtml(assessments[0].estimatedTimeline)}</p>`); | ||
|
|
||
| if (allObstacles.size > 0) { | ||
| const obstacleList = [...allObstacles].slice(0, MAX_IMPLEMENTATION_OBSTACLES).map(o => `<li>${escapeHtml(o)}</li>`).join(''); | ||
| parts.push(` <p><strong>Key obstacles:</strong></p>\n <ul>${obstacleList}</ul>`); | ||
| } | ||
|
|
||
| if (allAgencies.size > 0) { | ||
| parts.push(` <p><strong>Agencies involved:</strong> ${[...allAgencies].slice(0, MAX_AGENCIES_DISPLAYED).map(a => escapeHtml(a)).join(', ')}</p>`); | ||
| } |
Comment on lines
159
to
167
| const { frameworkAnalysis, perspectiveAnalysis } = analyzeDocumentsForContent(propositions, lang, data.ciaContext); | ||
| content += generateDeepAnalysisSection({ | ||
| documents: propositions, | ||
| lang, | ||
| cia: data.ciaContext, | ||
| articleType: 'propositions', | ||
| frameworkAnalysis, | ||
| perspectiveAnalysis, | ||
| }); |
Comment on lines
+908
to
+910
| const impacts = [...impactMap.values()].slice(0, MAX_STAKEHOLDER_IMPACTS); | ||
| if (impacts.length === 0) return ' <p>No stakeholder impact data available.</p>'; | ||
|
|
Comment on lines
+923
to
+935
| /** | ||
| * Render a risk assessment summary. Groups risks by type and shows up to 5. | ||
| */ | ||
| function renderRiskAssessment(risks: RiskAssessment[]): string { | ||
| // Deduplicate by type+severity, preferring higher severity | ||
| const byType = new Map<string, RiskAssessment>(); | ||
| for (const r of risks) { | ||
| const key = r.type; | ||
| const existing = byType.get(key); | ||
| if (!existing || severityRank(r.severity) > severityRank(existing.severity)) { | ||
| byType.set(key, r); | ||
| } | ||
| } |
Comment on lines
+962
to
+973
|
|
||
| for (const ia of assessments) { | ||
| ia.keyObstacles.forEach(o => allObstacles.add(o)); | ||
| ia.agenciesInvolved.forEach(a => allAgencies.add(a)); | ||
| if (feasibilityRank(ia.feasibility) < feasibilityRank(highestFeasibility)) { | ||
| highestFeasibility = ia.feasibility; | ||
| } | ||
| } | ||
|
|
||
| const parts: string[] = []; | ||
| const fIcon = highestFeasibility === 'high' ? '🟢' : highestFeasibility === 'medium' ? '🟡' : '🔴'; | ||
| parts.push(` <p>${fIcon} <strong>Feasibility:</strong> ${escapeHtml(highestFeasibility)}. ${escapeHtml(assessments[0].estimatedTimeline)}</p>`); |
Comment on lines
130
to
138
| const { frameworkAnalysis, perspectiveAnalysis } = analyzeDocumentsForContent(motions, lang, data.ciaContext); | ||
| content += generateDeepAnalysisSection({ | ||
| documents: motions, | ||
| lang, | ||
| cia: data.ciaContext, | ||
| articleType: 'motions', | ||
| frameworkAnalysis, | ||
| perspectiveAnalysis, | ||
| }); |
Comment on lines
89
to
97
| const { frameworkAnalysis, perspectiveAnalysis } = analyzeDocumentsForContent(interpellations, lang, data.ciaContext); | ||
| content += generateDeepAnalysisSection({ | ||
| documents: interpellations, | ||
| lang, | ||
| cia: data.ciaContext, | ||
| articleType: 'interpellations', | ||
| frameworkAnalysis, | ||
| perspectiveAnalysis, | ||
| }); |
Member
|
@copilot apply changes based on the comments in this thread |
…nderers Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>
Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>
Contributor
Author
Implemented the review-thread fixes in 881b7ff:
Screenshot of updated deep-analysis UI: |
Contributor
🔍 Lighthouse Performance Audit
📥 Download full Lighthouse report Budget Compliance: Performance budgets enforced via |
Contributor
🔍 Lighthouse Performance Audit
📥 Download full Lighthouse report Budget Compliance: Performance budgets enforced via |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
analyzeDocumentsForContent()when deep analysis will not render (doc count < 2 for standard types)analysis-integration,data-transformers) and full suite (3730/3730pass)tsconfig.scripts.json) with no errorsOriginal prompt
This section details on the original issue you should resolve
<issue_title>AI-Powered Comprehensive Document Analysis Framework with Multi-Stakeholder Perspectives</issue_title>
<issue_description>## 📋 Issue Type
Feature / Architecture Enhancement
🎯 Objective
Create an AI-powered Comprehensive Document Analysis Framework that ensures every parliamentary document, government proposition, and policy paper is systematically analyzed from all stakeholders' perspectives using structured analytical techniques (SWOT, PESTLE, stakeholder mapping, impact assessment) with multiple AI iterations for depth and accuracy. This framework becomes the shared analytical backbone that all content generators and agentic workflows call for consistent, high-quality political analysis.
📊 Current State
generateArticleContent()formats document metadata (title, type, date) — not the document's actual policy contentpolicy-analysis.ts: ContainsdetectPolicyDomains()(keyword matching),assessConfidenceLevel()(heuristic),detectNarrativeFrames()(pattern matching) — useful utilities but not AI-powered analysisrisk-analysis.ts: ContainscalculateCoalitionRiskIndex()anddetectAnomalousPatterns()— formula-based, not AI-analyzed🚀 Desired State
Comprehensive Analysis Framework
Every document processed through the framework receives:
Multi-Iteration Analysis Protocol
Stakeholder Universe (Dynamic Selection)
AI selects relevant stakeholders per document from this universe:
🔧 Implementation Approach
Step 1: Create Core Analysis Framework