🗂️ Analysis Artifact Catalog — Riksdagsmonitor

📊 Single Source of Truth for Every Markdown Artifact Produced by an Agentic News Workflow
🎯 One row per artifact · Methodology + Template + Depth Floor + Mermaid Type + MCP Data Sources

📋 Document Owner: CEO | 📄 Version: 1.3 | 📅 Last Updated: 2026-05-01 (UTC) 🔄 Review Cycle: Quarterly | ⏰ Next Review: 2026-07-31 🏢 Owner: Hack23 AB (Org.nr 5595347807) | 🏷️ Classification: Public

---

🎯 AI-FIRST Methodology Card

🚦 Read this card before writing a single paragraph. It names the artifact this methodology owns, the gate check it satisfies, the evidence-density target it must hit, and the Pass-1 / Pass-2 discipline required by .github/copilot-instructions.md §5 (AI-FIRST Quality Principle).

Field	Value
Purpose	Authoritative row-per-artifact catalog — every markdown artifact an agentic workflow can produce, with its family, methodology, template, depth floor, Mermaid type, and MCP data sources.
Inputs	reference-quality-thresholds.json; every Family methodology file; analysis/templates/; .github/prompts/04-analysis-pipeline.md
Outputs	(reference catalog — no daily artifact)
Owning artifact(s)	(documents all 23 always-on + Family E + supplementary)
Owning gate check	Check 1 (artifact existence) and Check 11 (supplementary) in 05-analysis-gate.md
Citation density target	Every catalog row cites the methodology + template file path (no narrative density target)
Banned phrases	Enforced via political-style-guide.md §Machine-readable banned-phrase list
Threshold source	reference-quality-thresholds.json → thresholds[articleType][artifact] (fallback defaults.coreArtifactFloor)

✅ Pass-1 checklist (creation — minimal viable artifact)

• Confirm every artifact filename in the catalog exists in analysis/templates/ (or is documented as agent-generated)
• Confirm every depth floor matches reference-quality-thresholds.json
• Produce every required sub-section listed in the owning template
• Add ≥ 1 evidence anchor (dok_id, vote id, named MP, or primary-source URL) per analytical claim
• Apply the correct WEP confidence band for the run's horizon (72h / week / month / quarter / year / cycle)
• Include ≥ 1 themed Mermaid diagram with style … or themeVariables config (where structurally meaningful)
• Cross-link the relevant template under analysis/templates/ and the gate check it satisfies

🔁 Pass-2 checklist (read-back & improve — AI-FIRST mandatory)

• Cross-walk: every catalog row → existing methodology section → existing template file → gate-check coverage
• Reject any row where methodology, template, or threshold is missing or stale
• Re-read the file end-to-end; flag every claim that lacks an evidence anchor and add one
• Replace every banned phrase listed in political-style-guide.md §Machine-readable banned-phrase list with an evidence-anchored alternative
• Tighten WEP language: never above likely without ≥ 3 cycle-aged sources for year/cycle horizons
• Strengthen Mermaid (color-coded style … directives, themeVariables, ≥ 5 nodes where the structure admits it)
• Add ≥ 1 second-order effect, cui-bono note, or counterfactual where the artifact admits one
• Verify citation density meets the per-file target below and the gate's evidence-density rules

🟢 Exemplar (good — pattern-match this)

(catalog row) synthesis-summary.md | Family A core | synthesis-methodology.md §3 | analysis/templates/intelligence-assessment.md | floor=205 (breaking) | Mermaid: flowchart | MCP: riksdag-regering, scb

🔴 Anti-exemplar (failure mode — never ship this)

(failure mode) synthesis | core | uses several methodologies | varies — no filename, no methodology link, no threshold reference, no Mermaid type, no MCP source.

🔗 Cross-links

• Template(s): Catalog rows reference every template under analysis/templates/
• Gate check: .github/prompts/05-analysis-gate.md
• AI-FIRST canon: .github/copilot-instructions.md §5 · ai-driven-analysis-guide.md
• Style canon: political-style-guide.md · osint-tradecraft-standards.md
• Catalog row: artifact-catalog.md

---

🎯 Purpose

This catalog is the authoritative index into every markdown artifact an agentic news workflow produces under analysis/daily/$ARTICLE_DATE/$SUBFOLDER/. For every artifact it names:

• the family (A / B / C / D / E) and whether it is a core always-on artifact or an operational supplementary enrichment,
• the methodology the AI agent applies when writing it,
• the template that defines output shape,
• the minimum line floor from reference-quality-thresholds.json,
• the mandatory color-coded Mermaid diagram type,
• the riksdag-regering MCP tools feeding it, and
• the gate check from .github/prompts/05-analysis-gate.md that enforces it.

Agents MUST read this catalog once at the start of every run, before opening any other methodology or template. It is referenced from ai-driven-analysis-guide.md, .github/prompts/04-analysis-pipeline.md and .github/prompts/05-analysis-gate.md.

---

🗺️ Artifact Group Overview

%%{init: {"theme":"dark","themeVariables":{"primaryColor":"#1565C0","primaryTextColor":"#ffffff","primaryBorderColor":"#0A3F7F","lineColor":"#90CAF9","secondaryColor":"#2E7D32","secondaryTextColor":"#ffffff","tertiaryColor":"#FF9800","tertiaryTextColor":"#000000","mainBkg":"#1565C0","secondBkg":"#2E7D32","tertiaryBkg":"#FF9800","noteBkgColor":"#FFC107","noteTextColor":"#000000","errorBkgColor":"#D32F2F","fontFamily":"Inter, Helvetica, Arial, sans-serif"}}}%%
flowchart LR
    subgraph INPUT["📥 Run Root"]
        R["analysis/daily/$ARTICLE_DATE/$SUBFOLDER/"]
    end
    subgraph CORE["🧠 Core Always-On — 24 artifacts"]
        A["Family A — Synthesis (9)"]
        B["Family B — Structural (2)"]
        C["Family C — Strategic (6)"]
        D["Family D — Electoral/Domain (7)"]
    end
    subgraph E["📑 Family E — Per-document (N)"]
        E1["documents/{dok_id}-analysis.md"]
    end
    subgraph SUPP["🛰️ Operational Supplementary"]
        S1["analysis-index.md"]
        S2["reference-analysis-quality.md"]
        S3["mcp-reliability-audit.md"]
        S4["workflow-audit.md"]
        S5["cross-run-diff.md"]
        S6["cross-session-intelligence.md"]
        S7["session-baseline.md"]
        S8["horizon-pir-rollforward.md"]
    end
    R --> CORE
    R --> E
    R --> SUPP
    CORE --> GATE["🚦 05-analysis-gate.md"]
    SUPP -. enrichment .-> GATE

    style CORE fill:#1565C0,color:#ffffff
    style E fill:#2E7D32,color:#ffffff
    style SUPP fill:#FF9800,color:#000000
    style GATE fill:#D32F2F,color:#ffffff

• Core always-on (24) — every run produces all 24, depth varies by tier (L1 / L2 / L2+ / L3) per the DIW Output Matrix. Missing any core artifact fails the gate.
• Family E — one per dok_id in the manifest; count is N = |manifest|. Low-weight items may be clustered.
• Operational supplementary (8) — enrichment artifacts that strengthen the AI-FIRST loop. They are recommended for deep and mandatory for comprehensive (Tier-C aggregation). They are not counted in the 24-artifact mandatory set but feed the self-correction, cross-run memory, and MCP health audit.

---

📘 Family A — Core Synthesis (9 artifacts · F3EAD: EXPLOIT → ANALYZE)

#	Canonical filename	Methodology §link	Template	Line floor (breaking)	Mermaid	Primary MCP tool
1	README.md	synthesis-methodology.md	templates/README.md § Folder README	40	—	run-internal
2	executive-brief.md	synthesis-methodology.md	executive-brief.md	90	flowchart	search_dokument + all
3	synthesis-summary.md	synthesis-methodology.md	synthesis-summary.md	205	mindmap	search_dokument + all
4	significance-scoring.md	synthesis-methodology.md	significance-scoring.md	150	flowchart rank	search_dokument, search_anforanden
5	classification-results.md	political-classification-guide.md	political-classification.md	140	flowchart	get_dokument
6	swot-analysis.md	political-swot-framework.md	swot-analysis.md	160	quadrantChart	get_voteringar, search_anforanden
7	risk-assessment.md	political-risk-methodology.md	risk-assessment.md	180	flowchart L×I	search_dokument, get_betankanden
8	threat-analysis.md	political-threat-framework.md	threat-analysis.md	180	attack-tree	search_anforanden, get_voteringar
9	stakeholder-perspectives.md	synthesis-methodology.md	stakeholder-impact.md	220	graph TB	search_ledamoter, get_ledamot

Filename variant — canonical stakeholder-perspectives.md ← template stakeholder-impact.md; canonical classification-results.md ← template political-classification.md.

---

📗 Family B — Structural Metadata (2 artifacts)

#	Canonical filename	Methodology §link	Template	Line floor	Mermaid	Primary source
10	data-download-manifest.md	structural-metadata-methodology.md	data-download-manifest.md	60	pie / bar	scripts in 03-data-download.md
11	cross-reference-map.md	structural-metadata-methodology.md	cross-reference-map.md	110	graph LR	sibling folders under $ARTICLE_DATE

Family B is scaffolded by scripts (no analytical prose) — agents only annotate confidence and data-depth tags.

---

📙 Family C — Strategic Extensions (6 artifacts · F3EAD: ANALYZE continued)

#	Canonical filename	Methodology §link	Template	Line floor	Mermaid	Primary MCP tool
12	scenario-analysis.md	strategic-extensions-methodology.md	scenario-analysis.md	180	flowchart + prob	search_voteringar, search_dokument. Horizon-stratified branches required when horizonDays >= 90
13	comparative-international.md	strategic-extensions-methodology.md	comparative-international.md	150	graph LR Nordic/EU	world-bank, scb, IMF
14	devils-advocate.md	strategic-extensions-methodology.md	devils-advocate.md	160	matrix (ACH)	cross-MCP
15	intelligence-assessment.md	strategic-extensions-methodology.md	intelligence-assessment.md	160	flowchart KJs	synthesis layer
16 ⭐	methodology-reflection.md	osint-tradecraft-standards.md	methodology-reflection.md	200	flowchart audit	run-internal
17	parliamentary-season.md	per-artifact-methodologies.md#parliamentary-season	parliamentary-season.md	160	gantt calendar	get_calendar_events, get_propositioner. Quarter-ahead + year-ahead + election-cycle only

⭐ methodology-reflection.md is the self-audit gate: evidence sufficiency, confidence distribution, source diversity, party-neutrality arithmetic, ICD 203 compliance, ≥ 3 concrete improvements for next run.

---

📕 Family D — Electoral & Domain Lenses (7 artifacts + 1 cycle-only)

#	Canonical filename	Methodology §link	Template	Line floor	Mermaid	Primary MCP tool
18	election-2026-analysis.md	electoral-domain-methodology.md	election-2026-analysis.md	150	bar + coalition	search_voteringar, SCB polls
19	voter-segmentation.md	electoral-domain-methodology.md	voter-segmentation.md	140	graph TB segments	SCB demographics
20	coalition-mathematics.md	electoral-domain-methodology.md	coalition-mathematics.md	135	flowchart seats	search_voteringar
21	historical-parallels.md	electoral-domain-methodology.md	historical-parallels.md	130	timeline	historical archive
22	media-framing-analysis.md	electoral-domain-methodology.md	media-framing-analysis.md	200	frame graph + laundering chain + lifecycle xychart + Outlet Bias Audit (5-axis) + Global Audience Orientation	external press (OSINT) + Outlet Bias Audit (Nordicom / Reuters Institute / Förvaltningsstiftelsen / PO-PON / EUvsDisinfo / DISARM TTPs) + international quality press + state-affiliated amplification monitoring
23	implementation-feasibility.md	electoral-domain-methodology.md	implementation-feasibility.md	140	flowchart risk	search_dokument (myndigheter)
24	forward-indicators.md	electoral-domain-methodology.md	forward-indicators.md	150	gantt 4-horizon	mixed. Horizon-band column required when horizonDays >= 90
25 🗳️	cycle-trajectory.md	per-artifact-methodologies.md#cycle-trajectory	cycle-trajectory.md	180	bar + timeline	SCB, IMF, search_voteringar. election-cycle only; blocking for cycle workflows

Filename variants — historical-parallels.md ↔ historical-baseline.md; comparative-international.md ↔ international-comparative.md; election-2026-analysis.md ↔ election-2026-implications.md. All map to a single template + methodology section per ai-driven-analysis-guide.md §Filename variants. Artifact filenames are the canonical identifiers; ordinal numbers in this catalog are catalog-local only and must not be hardcoded in template or prose text. Refer to cycle-trajectory.md by filename rather than by phrases such as "24th artifact." cycle-trajectory.md is produced only by election-cycle workflows.

---

📒 Family E — Per-Document (N artifacts)

One file per dok_id in data-download-manifest.md, written under documents/{dok_id}-analysis.md using per-file-political-intelligence.md at the depth tier matching the document's DIW score (L1 Surface / L2 Strategic / L2+ Priority / L3 Intelligence-grade).

• Cluster files — {cluster}-cluster-analysis.md collapse related low-weight items (e.g. 12 procedural committee notices → one cluster analysis).
• Methodology: per-document-methodology.md.
• Per-file floor is adaptive: max(DIW tier floor, --min-lines). See reference-quality-thresholds.json §defaults.perDocumentFloorByTier.

---

🛰️ Operational Supplementary Artifacts (8)

These artifacts are not counted in the 24-artifact mandatory set but strengthen quality, cross-run memory, MCP health and workflow-self-audit. They are:

• Recommended for every deep run.
• Mandatory for every comprehensive (Tier-C aggregation) run.
• Enforced by 05-analysis-gate.md §Supplementary checks when present; for non-Tier-C runs, absence generally does not fail the gate.
• Exception: S5 / cross-run-diff.md becomes gate-required whenever ANALYSIS_RUN_COUNT >= 2 (same article type), including standard and deep runs — the gate treats multi-run comparison as a required supplementary check independent of tier.

#	Canonical filename	Methodology §link	Template	Line floor	Purpose
S1	analysis-index.md	per-artifact-methodologies.md#analysis-index	analysis-index.md	120	Read-me-first index naming every artifact in the run + recommended reading order
S2	reference-analysis-quality.md	per-artifact-methodologies.md#reference-analysis-quality	reference-analysis-quality.md	120	Per-run self-score vs reference benchmark; Pass-2 action list
S3	mcp-reliability-audit.md	per-artifact-methodologies.md#mcp-reliability-audit	mcp-reliability-audit.md	150	Endpoint-by-endpoint riksdag-regering / scb / world-bank / IMF health record
S4	workflow-audit.md	per-artifact-methodologies.md#workflow-audit	workflow-audit.md	120	Prompt-module-by-module self-audit: which phases ran, where rules slipped
S5	cross-run-diff.md	per-artifact-methodologies.md#cross-run-diff	cross-run-diff.md	130	Bayesian delta vs previous run of same article type
S6	cross-session-intelligence.md	per-artifact-methodologies.md#cross-session-intelligence	cross-session-intelligence.md	140	Session-over-session progression (week / month / quarter aggregation)
S7	session-baseline.md	per-artifact-methodologies.md#session-baseline	session-baseline.md	140	Calendar + adopted-texts + votering roster for the period
S8	horizon-pir-rollforward.md	per-artifact-methodologies.md#horizon-pir-rollforward	horizon-pir-rollforward.md	100	PIR genealogy across long-horizon runs. Never blocking; recommended when predecessor exists

---

🔭 Analytical Supplementary Artifacts (4)

Optional deep-dive templates mapped to analytical frameworks explicitly listed in the intelligence-operative agent's Core Expertise but previously untemplated. Never replace a core artifact and never blocking in 05-analysis-gate.md. Full production rules: analytical-supplementary-methodology.md.

#	Canonical filename	Methodology §link	Template	Line floor (standard / deep / comp)	Trigger
AS1	pestle-analysis.md	analytical-supplementary-methodology.md#pestle	pestle-analysis.md	100 / 150 / 220	event crosses ≥ 2 PESTLE dimensions
AS2	political-stride-assessment.md	analytical-supplementary-methodology.md#stride-political	political-stride-assessment.md	110 / 160 / 240	election-adjacent · integrity incident · disinfo spike · critical-infra vote
AS3	wildcards-blackswans.md	analytical-supplementary-methodology.md#wildcards--black-swans	wildcards-blackswans.md	110 / 160 / 240	long-horizon forecasting (monthly-review, election-year aggregation)
AS4	quantitative-swot.md	analytical-supplementary-methodology.md#quantitative-swot	quantitative-swot.md	110 / 160 / 240	decision memo requiring scored ranking

---

🧠 MCP Data Source Inventory

Server	Transport	Key tools	Feeds artifacts
riksdag-regering	HTTP	search_ledamoter, get_ledamot, search_dokument, get_dokument, search_anforanden, search_voteringar, get_voteringar, get_calendar_events, get_betankanden, get_propositioner, get_motioner, get_fragor, get_interpellationer	1–11, 14, 17–25, E-files
scb	local PxWeb v2	table-search + query	13, 18–20, 23, 25, S7
world-bank	local	get-social/health/education/environment	13, 23, S7
Statskontoret (web_fetch)	public web	agency-capacity reports, public-management evaluations, administrative burden evidence	7, 9, 13, 22, S1/S3/S7
IMF (tsx scripts/imf-fetch.ts)	CLI	WEO + SDMX 3.0 passthrough	13, 18, 23, 25, S7
github	HTTP Insiders	full toolset	PR creation (module 07)
filesystem / memory / sequential-thinking / playwright	local	helpers	cross-cutting

See .github/copilot-mcp.json for the canonical definition and 02-mcp-access.md for access rules.

---

🚦 Gate-Check Mapping (conceptual — see 05-analysis-gate.md for canonical numbering)

Note: The table below is a conceptual mapping of gate concerns to artifact scope. The authoritative check numbering and semantics live in .github/prompts/05-analysis-gate.md; the numbers here are illustrative groupings, not 1:1 references to the script's # Check N labels.

Gate concern	Artifacts audited
Artifact presence	1–24 (+ 25 for election-cycle)
Family-E per-dok_id coverage	E files
No-stub / depth floor per tier	1–24 (+ 25 for election-cycle) + E (via reference-quality-thresholds.json)
Evidence citations (dok_id, vote counts, named actors, Admiralty grade)	All prose
Mermaid diagram presence (color-coded)	1–24 (+ 25 for election-cycle) where row specifies Mermaid
Pass-2 evidence (mtime / diff)	Core prose files
Family C structure	12–16 (KJ count, comparator rows, etc.)
Family D structure	18–25 where applicable
PIR status sidecar	pir-status.json
Supplementary (Tier-C only)	S1–S8
Long-horizon additive (quarter/year/cycle)	LH checks per ext/long-horizon-forecasting.md

---

🔄 Maintenance

• When an artifact is added, renamed, or removed, update: this catalog, reference-quality-thresholds.json, per-artifact-methodologies.md, the template README, and .github/prompts/04-analysis-pipeline.md. All five must stay in lock-step.
• When a methodology section materially changes, increment the methodology's minor version and note it here under "Changelog".
• The depth floor benchmark is re-calibrated every quarter from the best-performing production run of the preceding quarter (the "reference run"). See reference-quality-thresholds.json §referenceBenchmark`.

📜 Changelog

• v1.3 (2026-05-01) — Added cycle-trajectory.md (Family D #25, election-cycle only, blocking), parliamentary-season.md (Family C #17, quarter/year/cycle), horizon-pir-rollforward.md (S8, never blocking). Appended horizon-band notes to forward-indicators.md and scenario-analysis.md rows. Renumbered Family D artifacts 17→18 through 23→24. Updated core count 23→24 (25 for cycle workflows).
• v1.1 (2026-04-23) — Added Analytical Supplementary Artifacts (4): pestle-analysis.md, political-stride-assessment.md, wildcards-blackswans.md, quantitative-swot.md. Linked to new analytical-supplementary-methodology.md. Non-blocking.
• v1.0 (2026-04-23) — Initial Riksdagsmonitor catalog extracted and adapted from EU Parliament Monitor artifact-catalog.md v1.0. 23 core artifacts + 7 supplementary + N per-document; aligned with ai-driven-analysis-guide.md v4.2.