synthesis-summary.md

May 3, 2026 · View on GitHub

🧩 Political Intelligence Synthesis Template

📊 Integrated Analysis Summary Combining All Intelligence Streams
🎯 Classification · SWOT · Risk · Threat · Stakeholder · Forward Outlook

📋 Document Owner: CEO | 📄 Version: 2.5 | 📅 Last Updated: 2026-04-25 (UTC)
🏢 Owner: Hack23 AB (Org.nr 5595347807) | 🏷️ Classification: Public

📌 Template Instructions: Copy to analysis/daily/YYYY-MM-DD/{articleType}/ and save as synthesis-summary.md in the workflow's own folder. This file synthesizes per-file analyses into an integrated intelligence picture. AI reads all per-file analyses and produces genuine synthesis — not a mechanical concatenation of summaries.

🚨 Anti-Pattern Warning: A synthesis that merely lists document titles without analytical connections is REJECTED. Every synthesis MUST:

Identify cross-document patterns (what themes emerge across multiple documents?)

Assess aggregate SWOT (combining individual SWOT findings using intersection rules)

Map risk interconnections (how do individual risk findings compound?)

Provide forward intelligence (what should we watch for next? Specific triggers.)

Include ≥2 color-coded Mermaid diagrams (intelligence dashboard + one other)

Rank documents by significance and explain the ranking rationale

📐 Template Contract — every fill of this template MUST satisfy this row.

Slot Value
Owning methodology per-artifact-methodologies.md
Owning gate check Check 1 + Check 5 (Mermaid) — see 05-analysis-gate.md
Required inputs all Family A peers + cross-reference-map.md
Horizon band per-run (per scripts/horizon-context.ts)
Output family Family A — Core Synthesis
Aggregation order 2 of 30 in canonical order (see scripts/render-lib/aggregator/order.ts)
Reader Intelligence Guide row generated from synthesis-summary.md (see scripts/render-lib/aggregator/reader-guide.ts)
Canonical evidence anchor | claim | evidence (dok_id / vote / MP intressent_id / primary-source URL) | retrieved_at | confidence | — every analytical claim row uses this schema.

Cross-reference: README.md §Template ↔ Methodology ↔ Gate-Check Matrix.

Slot	Value
Owning methodology	`per-artifact-methodologies.md`
Owning gate check	Check 1 + Check 5 (Mermaid) — see `05-analysis-gate.md`
Required inputs	all Family A peers + cross-reference-map.md
Horizon band	per-run (per `scripts/horizon-context.ts`)
Output family	Family A — Core Synthesis
Aggregation order	2 of 30 in canonical order (see `scripts/render-lib/aggregator/order.ts`)
Reader Intelligence Guide	row generated from `synthesis-summary.md` (see `scripts/render-lib/aggregator/reader-guide.ts`)
Canonical evidence anchor	`\| claim \| evidence (dok_id / vote / MP intressent_id / primary-source URL) \| retrieved_at \| confidence \|` — every analytical claim row uses this schema.

🔄 Tradecraft Context

Element	Value
F3EAD Stage	ANALYZE → DISSEMINATE — integrates per-document Family E findings into the run's coherent intelligence picture, identifies cross-document patterns, and lights the path from raw documents to executive-brief BLUF.
PIRs Served	Serves all standing PIRs that the day's documents touch; the synthesis explicitly lists which of PIR-1..PIR-7 each major finding addresses, and surfaces any PIR with insufficient evidence.
Admiralty Floor	B2 floor on synthesis-level claims; per-document anchor evidence inherits the Admiralty grade from the contributing Family E `{dok_id}-analysis.md`; aggregate confidence ≤ floor of contributing artifact confidences.
WEP + ODNI	Pattern statements use WEP phrasing for cross-document inferences (`a likely coordinated push`, `about even between two interpretations`); aggregate confidence label is the floor (not average) of contributing-artifact confidences.
Source Diversity Floor	The synthesis must integrate ≥3 different MCP sources (riksdag-regering MCP + at least one of `data.scb.se`, IMF, regeringen.se / `data.regeringen.se`); a synthesis built from a single MCP source is downgraded to ≤ MEDIUM confidence regardless of within-source breadth.
SAT(s) Applied	Cross-Impact Analysis (between same-day documents); Pattern Recognition (against the same-type 30-day baseline + same-type cross-run-diff); ACH lite (when ≥2 competing narratives explain the day); Indicators & Signposts (forward triggers extracted to `forward-indicators.md`).
ICD 203 Standards	All 9 standards apply — synthesis is the artifact most directly evaluated against ICD 203; in particular 4 (relevance to consumer), 5 (sourcing), 6 (logical argumentation), 7 (uncertainty), 8 (analytic value), 9 (alternative analysis).

See osint-tradecraft-standards.md for canonical F3EAD / PIR catalogue / Admiralty Code / WEP / SAT / ICD 203 definitions, and synthesis-methodology.md for DIW-weighted ranking, lead-story selection, and aggregate confidence rules.

📋 Synthesis Context

Field	Value
Synthesis ID	`[REQUIRED: SYN-YYYY-MM-DD-NNN]`
Analysis Date	`[REQUIRED: YYYY-MM-DD HH:MM UTC]`
Documents Analyzed	`[REQUIRED: N]`
Analysis Period	`[REQUIRED: e.g. "2026-03-28 00:00–18:00 UTC"]`
Produced By	`[REQUIRED: workflow name]`
Overall Confidence	`[REQUIRED: VERY HIGH / HIGH / MEDIUM / LOW / VERY LOW]`

📊 Data Quality Assessment

AI Instructions: Count the data depth of ALL documents analyzed. Classify FULL-TEXT using the actual full-text fields (fullText or fullContent) based on presence and meaningful length; treat contentFetched only as an auxiliary signal that content retrieval occurred, not as proof of full text. Classify SUMMARY-ONLY when no fullText/fullContent is present but a substantive summary/notis exists. Classify METADATA-ONLY when neither full text nor substantive summary is present. This assessment determines the maximum permissible confidence level for the entire synthesis. If the majority of documents are METADATA-ONLY, the synthesis confidence MUST NOT exceed MEDIUM.

Metric	Value
Documents with full text (`fullText`/`fullContent` present with meaningful content; `contentFetched` auxiliary only)	`[REQUIRED: X of Y]`
Documents with summary only (no `fullText`/`fullContent`; substantive summary/notis only, e.g. 100-500 chars)	`[REQUIRED: X of Y]`
Documents metadata-only (no `fullText`/`fullContent` and no substantive summary; title/date/committee only)	`[REQUIRED: X of Y]`
Maximum permissible confidence	`[REQUIRED: Based on ratio — if >50% metadata-only → max MEDIUM; if >75% metadata-only → max LOW]`
Data enrichment method	`[REQUIRED: e.g. "download-parliamentary-data enrichment", "AI direct MCP calls", "metadata-only"]`

⚠️ Confidence Override Rule: The Overall Confidence in the Synthesis Context table above MUST NOT exceed the Maximum permissible confidence calculated here. If it does, reduce the Overall Confidence to match.

📊 Intelligence Dashboard

Daily Political Landscape

AI Instructions: Replace all placeholder values with actual analysis results. Update each node's style line from grey dashed placeholder to the appropriate level color:

Sensitivity: 🟢 PUBLIC #4CAF50 · 🟡 SENSITIVE #FFC107 · 🔴 RESTRICTED #D32F2F

Risk / Threat / Significance: use the standard palette (#D32F2F / #FF9800 / #FFC107 / #4CAF50)

graph TD
    subgraph "📊 Daily Political Intelligence Dashboard"
        direction TB
        subgraph "🔒 Sensitivity"
            CLS["Sensitivity<br/>[PUBLIC/SENSITIVE/RESTRICTED]"]
        end
        subgraph "⚖️ Risk"
            RSK["Overall Risk<br/>[CRITICAL/HIGH/MEDIUM/LOW]<br/>Top: [RSK-NNN description]"]
        end
        subgraph "🎭 Threat"
            THR["Threat Level<br/>[SEVERE/HIGH/MODERATE/LOW]<br/>Top Category: [NI/LI/AC/TR/DP/PB]"]
        end
        subgraph "📈 Significance"
            SIG["Top Significance<br/>[#.#]/10<br/>[Breaking/Priority/Publish/Monitor]"]
        end
    end

    subgraph "🎯 Editorial Decision"
        DEC{Article Decision}
        DEC -->|"High urgency"| BRK["⚡ Breaking Article"]
        DEC -->|"Significant"| STD["📰 Standard Article"]
        DEC -->|"Low significance"| MON["📋 Analysis Only"]
    end

    CLS --> DEC
    RSK --> DEC
    THR --> DEC
    SIG --> DEC

    style CLS fill:#9E9E9E,stroke:#424242,stroke-width:2px,stroke-dasharray:5 5,color:#FFFFFF
    style RSK fill:#9E9E9E,stroke:#424242,stroke-width:2px,stroke-dasharray:5 5,color:#FFFFFF
    style THR fill:#9E9E9E,stroke:#424242,stroke-width:2px,stroke-dasharray:5 5,color:#FFFFFF
    style SIG fill:#9E9E9E,stroke:#424242,stroke-width:2px,stroke-dasharray:5 5,color:#FFFFFF
    style BRK fill:#D32F2F,color:#FFFFFF
    style STD fill:#4CAF50,color:#FFFFFF
    style MON fill:#9E9E9E,color:#FFFFFF

🏆 Top Findings by Significance

Rank	dok_id	Title	Significance	Risk Tier	SWOT Impact	Recommendation
1	`[REQUIRED]`	`[REQUIRED]`	`[#.#]`	`[🟢/🟡/🟠/🔴]`	`[S/W/O/T dominant]`	`[Breaking/Priority/Publish/Monitor]`
2	`[REQUIRED]`	`[REQUIRED]`	`[#.#]`	`[tier]`	`[quadrant]`	`[action]`
3	`[REQUIRED]`	`[REQUIRED]`	`[#.#]`	`[tier]`	`[quadrant]`	`[action]`
4	`[OPTIONAL]`	`[OPTIONAL]`	`[#.#]`	`[tier]`	`[quadrant]`	`[action]`
5	`[OPTIONAL]`	`[OPTIONAL]`	`[#.#]`	`[tier]`	`[quadrant]`	`[action]`

📖 Narrative (v3.2 — required)

Purpose: the table above is the analytic ranking; this section is the prose handoff to article.md for the lead story. Apply political-style-guide.md §"Narrative-Voice Standards" — choose one canonical lede pattern, name three people in the first 200 words, follow the sentence-cadence rule, include sensory specificity, close with a counter-narrative paragraph. Pass-2 grades this on the 6-axis narrative rubric.

Lead-story narrative (400–700 words)

Lede (one of: hard-news / tension-contrast / scene-setting / significance-first — pick the form that fits the day's #1 ranked finding)

[REQUIRED]

Body (2–4 paragraphs)

[REQUIRED — vary sentence length, name ≥ 3 actors with role + party + verb, ≥ 1 concrete sensory detail per 400 words. Tradecraft jargon allowed only with payoff within ≤ 2 sentences.]

Counter-narrative (60–150 words, signposted)

"There is a contrary read." [REQUIRED — named source whose framing differs.]

Secondary thread narrative (200–400 words, optional but recommended for ≥ 3-finding days)

[OPTIONAL — same structure, applied to ranking row #2 if its DIW is within 1.5 of #1.]

💪 Aggregated SWOT Summary

Combines individual document SWOT analyses into a landscape-level view.

Coalition Balance

graph LR
    subgraph "🏛️ Government Coalition Assessment"
        GS["✅ Strengths<br/>[N entries]<br/>Dominant: [summary]"]
        GW["⚠️ Weaknesses<br/>[N entries]<br/>Critical: [summary]"]
        GO["🚀 Opportunities<br/>[N entries]<br/>Top: [summary]"]
        GT["🔴 Threats<br/>[N entries]<br/>Top: [summary]"]
    end

    GS -.->|"exploits"| GO
    GW -.->|"amplifies"| GT
    GO -.->|"mitigates"| GW

    style GS fill:#4CAF50,color:#FFFFFF
    style GW fill:#FF9800,color:#FFFFFF
    style GO fill:#1565C0,color:#FFFFFF
    style GT fill:#D32F2F,color:#FFFFFF

Quadrant	Count	Highest-Impact Entry	Evidence
✅ Strengths	`[N]`	`[REQUIRED: strongest finding]`	`[dok_id]`
⚠️ Weaknesses	`[N]`	`[REQUIRED: most critical weakness]`	`[dok_id]`
🚀 Opportunities	`[N]`	`[REQUIRED: best opportunity]`	`[dok_id]`
🔴 Threats	`[N]`	`[REQUIRED: most serious threat]`	`[dok_id]`

SWOT Balance Assessment: [REQUIRED: 1–2 sentences — e.g. "Coalition strengths outweigh weaknesses this period, but electoral threat from S welfare narrative creates medium-term vulnerability."]

⚖️ Risk Landscape Summary

Risk Category	Score Range	Highest Risk	Trend vs. Previous
Coalition Stability	`[N–N]`	`[RSK-NNN: description]`	`[↑/→/↓]`
Policy Implementation	`[N–N]`	`[RSK-NNN: description]`	`[↑/→/↓]`
Budget / Fiscal	`[N–N]`	`[RSK-NNN: description]`	`[↑/→/↓]`
Electoral	`[N–N]`	`[RSK-NNN: description]`	`[↑/→/↓]`
Democratic Process	`[N–N]`	`[RSK-NNN: description]`	`[↑/→/↓]`
External / International	`[N–N]`	`[RSK-NNN: description]`	`[↑/→/↓]`

Overall Risk Level: [REQUIRED: LOW / MEDIUM / HIGH / CRITICAL]

🎭 Threat Summary

Threat Category	Threat Level	Key Finding
NI — Narrative Integrity	`[LOW/MOD/HIGH/SEVERE]`	`[1 sentence]`
LI — Legislative Integrity	`[LOW/MOD/HIGH/SEVERE]`	`[1 sentence]`
AC — Accountability	`[LOW/MOD/HIGH/SEVERE]`	`[1 sentence]`
TR — Transparency	`[LOW/MOD/HIGH/SEVERE]`	`[1 sentence]`
DP — Democratic Process	`[LOW/MOD/HIGH/SEVERE]`	`[1 sentence]`
PB — Power Balance	`[LOW/MOD/HIGH/SEVERE]`	`[1 sentence]`

Overall Threat Level: [REQUIRED: LOW / MODERATE / HIGH / SEVERE]

👥 Stakeholder Impact Overview

Stakeholder	Impact	Direction	Key Driver
🏘️ Citizens	`[VH/H/M/L/VL/N]`	`[positive/negative/neutral]`	`[REQUIRED]`
🏛️ Government	`[VH/H/M/L/VL/N]`	`[positive/negative/neutral]`	`[REQUIRED]`
🗳️ Opposition	`[VH/H/M/L/VL/N]`	`[positive/negative/neutral]`	`[REQUIRED]`
🏭 Business	`[VH/H/M/L/VL/N]`	`[positive/negative/neutral]`	`[REQUIRED]`
🤝 Civil Society	`[VH/H/M/L/VL/N]`	`[positive/negative/neutral]`	`[REQUIRED]`
🌍 International	`[VH/H/M/L/VL/N]`	`[positive/negative/neutral]`	`[REQUIRED]`

🎯 Narrative Direction & Article Decision

⚠️ v5.0 — ANALYSIS-DRIVEN ARTICLE DECISION: This section is the SINGLE SOURCE OF TRUTH for article generation. The AI MUST complete ALL analysis (SWOT, risk, threat, stakeholder, significance scoring, per-file analysis) BEFORE filling in this section. Article titles, descriptions, and SEO metadata are derived from these fields — NEVER from code templates or string interpolation.

[REQUIRED: 4–6 sentences providing the primary narrative direction for article generation. This is the lede thesis that the article generator should use. Be specific about the central political tension, the key actors, and the intelligence-level insight. Include confidence assessment.]

Primary Narrative Angle: [REQUIRED: 1 sentence — the article headline thesis]
Secondary Angles: [OPTIONAL: 1–2 alternative narrative framings]
Confidence: [REQUIRED: VERY HIGH / HIGH / MEDIUM / LOW / VERY LOW]

📰 AI-Recommended Article Metadata (MANDATORY — v5.0)

SEO contract: these fields must agree with executive-brief.md H1 + BLUF and satisfy .github/prompts/seo-metadata-contract.md. Do not use dates, boilerplate, passive noun phrases, or generic AI-generated political intelligence wording.

These fields MUST be generated by AI from the completed analysis above — NEVER from code templates. The article generator reads these fields from synthesis-summary.md and uses them directly for <title>, <meta name="description">, Schema.org headline and alternativeHeadline.

Field	Value	Requirements
Recommended Title (EN)	`[REQUIRED: 60-80 chars, newsworthy, names actors/institutions, uses active verbs]`	Must reference specific findings from Top Findings table above
Recommended Title (SV)	`[REQUIRED: Swedish equivalent, 60-80 chars]`	Professional Swedish political language
Meta Description (EN)	`[REQUIRED: 150-160 chars summarizing key intelligence findings]`	Must name specific policy areas, actors, and significance
Meta Description (SV)	`[REQUIRED: 150-160 chars]`	Swedish equivalent with proper political terminology
Key Highlights	`[REQUIRED: 3-5 specific, concrete highlights from analysis — NOT metadata labels]`	Each highlight: named actor + policy action + significance
Article Decision	`[REQUIRED: PUBLISH / ANALYSIS-ONLY / SKIP]`	Based on significance scores and risk levels above
Article Priority	`[REQUIRED: BREAKING / PRIORITY / STANDARD / MONITOR]`	Derived from highest significance score and risk tier
Justification	`[REQUIRED: 1-2 sentences explaining why this merits publication at the stated priority]`	Evidence-based, citing specific dok_ids and scores

BANNED Patterns in Recommended Metadata

❌ "Government Propositions: Policy Priorities This Week: {Topic} in Focus" — template pattern
❌ "Political intelligence analysis of N documents..." — document count subtitle
❌ "Analysis of N documents covering {Field}:, {Field}:" — metadata field leak
❌ Any title/description that could be generated by string interpolation without reading analysis
❌ Repeating the article type name as the title (e.g., "Committee Reports" as title)

✅ Quality Indicators for Recommended Metadata

Title names at least one specific actor, institution, or policy measure from the analysis
Meta description references the #1 ranked finding from the Top Findings table
Key highlights are unique, substantive facts — not section headers or generic categories
Article decision is justified by specific significance scores ≥ 5.0 (Publish) or ≥ 7.0 (Priority/Breaking)

📊 Historical Comparison

AI Instructions: Compare current period findings with equivalent periods from previous riksmöten. This establishes trend context and prevents over-indexing on routine developments.

Metric	Current Period	Prior Period (3 months ago)	Year Ago	Trend
Overall Risk Level	`[REQUIRED]`	`[OPTIONAL]`	`[OPTIONAL]`	`[↑/→/↓]`
Coalition Stability	`[REQUIRED]`	`[OPTIONAL]`	`[OPTIONAL]`	`[↑/→/↓]`
Legislative Throughput	`[REQUIRED: N documents]`	`[OPTIONAL]`	`[OPTIONAL]`	`[↑/→/↓]`
Opposition Activity Level	`[REQUIRED: VH/H/M/L/VL]`	`[OPTIONAL]`	`[OPTIONAL]`	`[↑/→/↓]`
Average Significance Score	`[REQUIRED: #.#/10]`	`[OPTIONAL]`	`[OPTIONAL]`	`[↑/→/↓]`

Historical Context: [REQUIRED: 2–3 sentences placing this period in historical context. Is this an unusually active/risky period? Compare to equivalent pre-election periods in 2021/22.]

Precedents from Prior Riksmöten:

Precedent	Riksmöte	Outcome	Relevance to Current Period
`[OPTIONAL: similar political situation]`	`[e.g. 2021/22]`	`[What happened]`	`[Why it matters now]`
`[OPTIONAL]`	`[year]`	`[outcome]`	`[relevance]`

🗳️ Election 2026 Implications

Dimension	Assessment	Evidence
Electoral Impact	`[REQUIRED: How does this affect September 2026 election positioning?]`	`[Specific evidence]`
Coalition Scenarios	`[REQUIRED: Which coalition configurations benefit/suffer?]`	`[Evidence]`
Voter Salience	`[REQUIRED: Which voter segments are most affected? By how much?]`	`[Evidence]`
Campaign Vulnerability	`[REQUIRED: Does this create campaign attack vectors for opposition?]`	`[Evidence]`
Policy Legacy	`[REQUIRED: Will this become an electoral asset or liability?]`	`[Evidence]`

Overall Electoral Significance: [REQUIRED: CRITICAL/HIGH/MODERATE/LOW/NEGLIGIBLE]

Most Likely Narrative: [REQUIRED: How will this be framed in the 2026 campaign?]

🔮 Forward Indicators (MANDATORY)

⚠️ This section is MANDATORY — analysis without forward indicators is incomplete and will be REJECTED.

#	Indicator	Timeline	Source	Watch Priority
1	`[REQUIRED: specific event or metric to monitor]`	`[days/weeks]`	`[data source]`	`🔴/🟠/🟡/🟢`
2	`[REQUIRED]`	`[timeline]`	`[source]`	`[tier]`
3	`[REQUIRED]`	`[timeline]`	`[source]`	`[tier]`

Aggregate Risk Level Summary:

Metric	Value	Trend vs. Previous
Overall Risk Level	`[REQUIRED: LOW / MEDIUM / HIGH / CRITICAL]`	`[↑/→/↓]`
Overall Threat Level	`[REQUIRED: LOW / MODERATE / HIGH / SEVERE]`	`[↑/→/↓]`
Highest Significance Score	`[REQUIRED: #.#/10]`	`[↑/→/↓]`
SWOT Balance	`[REQUIRED: Positive / Neutral / Negative]`	`[↑/→/↓]`

Previous Synthesis Reference: [REQUIRED: path to previous synthesis-summary.md or "N/A — first synthesis"]

📋 Analysis Artifacts Inventory

File	Status	Key Output
`classification-results.md`	`[✅/⚠️/❌]`	`[REQUIRED: main classification finding]`
`risk-assessment.md`	`[✅/⚠️/❌]`	`[REQUIRED: overall risk level]`
`swot-analysis.md`	`[✅/⚠️/❌]`	`[REQUIRED: SWOT balance]`
`threat-analysis.md`	`[✅/⚠️/❌]`	`[REQUIRED: overall threat level]`
`stakeholder-perspectives.md`	`[✅/⚠️/❌]`	`[REQUIRED: highest-impact stakeholder]`
`significance-scoring.md`	`[✅/⚠️/❌]`	`[REQUIRED: top significance score]`
Per-file `.analysis.md` files	`[N created]`	`[REQUIRED: count of per-file analyses]`

📂 MCP Data Files Used

[REQUIRED: List all MCP data file paths consulted for this synthesis. Include riksdag-regering-mcp tool outputs, CIA data exports, and any cached data files used during analysis.]

#	Data Source	File / Tool Path	Retrieved
1	`[e.g. riksdag-regering-mcp]`	`[e.g. search_dokument(doktyp="prop", rm="2025/26")]`	`[YYYY-MM-DD HH:MM UTC]`
2	`[e.g. CIA export]`	`[e.g. cia-data/exports/risk-summary.json]`	`[YYYY-MM-DD HH:MM UTC]`
3	`[e.g. riksdag-regering-mcp]`	`[e.g. search_voteringar(rm="2025/26")]`	`[YYYY-MM-DD HH:MM UTC]`
4	`[OPTIONAL]`	`[path or tool call]`	`[timestamp]`
5	`[OPTIONAL]`	`[path or tool call]`	`[timestamp]`

📌 AI Instructions: Populate this table with every MCP tool call and data file actually consulted during the synthesis workflow. This provides full data provenance and audit trail for the intelligence product.

🔗 Cross-References

Link to same-day analysis from other article types and related external intelligence products.

Related Analysis	Article Type	Date	Key Finding
`[OPTIONAL: e.g. analysis/daily/2026-04-04/propositions/synthesis-summary.md]`	`[propositions]`	`[date]`	`[1 sentence]`
`[OPTIONAL: e.g. analysis/daily/2026-04-04/committee-reports/synthesis-summary.md]`	`[committee-reports]`	`[date]`	`[1 sentence]`
`[OPTIONAL: CIA platform data]`	`[cia-export]`	`[date]`	`[1 sentence]`

🎯 Confidence Scale Reference (5-Level)

Level	Label	Criteria	Evidence Threshold
⬛ 1	VERY LOW	Speculation only, single unverified source	0–1 sources, no corroboration
🟥 2	LOW	Circumstantial evidence, indirect indicators	2 sources, indirect evidence
🟧 3	MEDIUM	Multiple independent sources, moderate corroboration	3+ sources, moderate agreement
🟩 4	HIGH	Official records, documented data, direct evidence	Official docs, voting records, committee reports
🟦 5	VERY HIGH	Verified data + independent corroboration + expert consensus	Multiple official sources, cross-validated

✅ Quality Self-Check Checklist

Pre-commit validation — every item MUST be checked before finalising this synthesis. Derived from the analysis-gate checks in .github/prompts/05-analysis-gate.md.

Document Control:

Template Path: /analysis/templates/synthesis-summary.md
Version: 2.5
Effective Date: 2026-04-25 (UTC)
Key Changes v2.3: Added Election 2026 Implications section, Historical Comparison tables, 5-level confidence scale (VERY HIGH/HIGH/MEDIUM/LOW/VERY LOW), updated quality checklist
Consumed By: All news article generator workflows
ISMS Alignment: ISO 27001:2022 A.5.7 (Threat Intelligence), NIST CSF 2.0 ID.RA (Risk Assessment)
Classification: Public
Owner: Hack23 AB (Org.nr 5595347807)
Next Review: 2026-06-30

✅ Pass-2 Self-Audit Checklist (v4.4 — required)

Purpose: AI-FIRST principle requires a Pass-2 read-back-and-improve. After producing this artifact in Pass 1, re-read it end-to-end and verify each item below. Document any remediation in methodology-reflection.md §"Pass-2 audit log". Any unchecked ❌ box at the end of Pass 2 forces a Pass-3 rewrite of the affected section.