Add color-coded Mermaid diagrams to all analysis templates and enhance agentic workflow instructions by Copilot · Pull Request #1444 · Hack23/riksdagsmonitor

Copilot · 2026-03-28T17:13:45Z

Address review rounds 1-11 (previous commits)
Address review round 12 (7 comments from 22da9a2):
- Update v1 → v2 quality-criteria refs in 4 more workflow files: news-article-generator.md, news-realtime-monitor.md, news-month-ahead.md, news-monthly-review.md
- Update SHARED_PROMPT_PATTERNS.md shared skill block v1 refs to v2 (political-analysis + quality-criteria; stakeholder-perspectives annotated as v1-only)
- Fix catalog id generation to use relative path within type dir (supports nested subdirs like worldbank/ind1/SE.json → id "ind1/SE")
- Expand per-file-intelligence-analysis.md Key Fields table: added Events, MPs, MCP Responses rows; clarified World Bank nested path structure
- Add regression test for nested type subdirectories with duplicate basenames (worldbank/ind1/SE.json + worldbank/ind2/SE.json)
All 14 catalog tests pass, 0 CodeQL alerts

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.

github-actions · 2026-03-28T17:14:26Z

🏷️ Automatic Labeling Summary

This PR has been automatically labeled based on the files changed and PR metadata.

Applied Labels: size-xs

Label Categories

🗳️ Content: news, dashboard, visualization, intelligence
💻 Technology: html-css, javascript, workflow, security
📊 Data: cia-data, riksdag-data, data-pipeline, schema
🌍 I18n: i18n, translation, rtl
🔒 ISMS: isms, iso-27001, nist-csf, cis-controls
🏗️ Infrastructure: ci-cd, deployment, performance, monitoring
🔄 Quality: testing, accessibility, documentation, refactor
🤖 AI: agent, skill, agentic-workflow

For more information, see .github/labeler.yml.

github-actions · 2026-03-28T17:14:50Z

🔍 Lighthouse Performance Audit

Category	Score	Status
Performance	85/100	🟡
Accessibility	95/100	🟢
Best Practices	90/100	🟢
SEO	95/100	🟢

📥 Download full Lighthouse report

Budget Compliance: Performance budgets enforced via budget.json

- Add catalog-downloaded-data.ts script + 11 tests - Add per-file-political-intelligence.md template (SWOT.md quality) - Add ai-driven-analysis-guide.md methodology - Add per-file-intelligence-analysis.md prompt (v2) - Update SHARED_PROMPT_PATTERNS.md with per-file block - Update news-evening-analysis.md with AI-driven analysis - Update all 10 other agentic workflow .md files Agent-Logs-Url: https://github.com/Hack23/riksdagsmonitor/sessions/21aec9bc-8b00-486e-bf37-be3a77e1d64b Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

github-actions · 2026-03-28T17:28:40Z

🔍 Lighthouse Performance Audit

Category	Score	Status
Performance	85/100	🟡
Accessibility	95/100	🟢
Best Practices	90/100	🟢
SEO	95/100	🟢

📥 Download full Lighthouse report

Budget Compliance: Performance budgets enforced via budget.json

Agent-Logs-Url: https://github.com/Hack23/riksdagsmonitor/sessions/21aec9bc-8b00-486e-bf37-be3a77e1d64b Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

github-actions · 2026-03-28T17:34:29Z

🔍 Lighthouse Performance Audit

Category	Score	Status
Performance	85/100	🟡
Accessibility	95/100	🟢
Best Practices	90/100	🟢
SEO	95/100	🟢

📥 Download full Lighthouse report

Budget Compliance: Performance budgets enforced via budget.json

- Fix analysis/README.md: remove duplicate section, add per-file analysis guide, update workflow integration, add conflict resolution, update script references - Improve political-swot-framework.md: add EXPIRED entry handling rules, landscape SWOT weighting (Gov 0.40, S 0.25, SD 0.20, minor 0.15), conflict resolution, minimum requirements, update pipeline to v2 - Improve political-risk-methodology.md: remove duplicate technical section, add AI analysis protocol, add Risk-to-SWOT integration rules, add dimension weighting, add calibration cross-references - Improve political-classification-guide.md: remove duplicate technical section, add borderline classification guidance, add score divergence protocol - Improve political-threat-framework.md: add severity calibration table (1-5), add escalation criteria, add Threat-to-Risk integration mapping, add AI analysis protocol with MCP tool references - Improve templates: add temporal decay tracking to SWOT template, add severity reference to threat template, add escalation decision table, add risk freshness/escalation to risk template, add score reconciliation to significance template, add conflicting impact resolution to stakeholder template, add calibration example to classification template, expand per-file template risk section to 6 categories with scoring reference Agent-Logs-Url: https://github.com/Hack23/riksdagsmonitor/sessions/a27cff28-2ee0-4999-be91-2724c85cb117 Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

github-actions · 2026-03-28T17:56:48Z

🔍 Lighthouse Performance Audit

Category	Score	Status
Performance	85/100	🟡
Accessibility	95/100	🟢
Best Practices	90/100	🟢
SEO	95/100	🟢

📥 Download full Lighthouse report

Budget Compliance: Performance budgets enforced via budget.json

…ication, L×I note, time budget guidance Agent-Logs-Url: https://github.com/Hack23/riksdagsmonitor/sessions/a27cff28-2ee0-4999-be91-2724c85cb117 Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

github-actions · 2026-03-28T18:02:32Z

🔍 Lighthouse Performance Audit

Category	Score	Status
Performance	85/100	🟡
Accessibility	95/100	🟢
Best Practices	90/100	🟢
SEO	95/100	🟢

📥 Download full Lighthouse report

Budget Compliance: Performance budgets enforced via budget.json

Agent-Logs-Url: https://github.com/Hack23/riksdagsmonitor/sessions/0088c258-7514-4a3f-b3bf-76d636f48db2 Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

github-actions · 2026-03-29T09:38:47Z

🔍 Lighthouse Performance Audit

Category	Score	Status
Performance	85/100	🟡
Accessibility	95/100	🟢
Best Practices	90/100	🟢
SEO	95/100	🟢

📥 Download full Lighthouse report

Budget Compliance: Performance budgets enforced via budget.json

Copilot · 2026-03-29T09:39:02Z

@copilot apply changes based on the comments in this thread

All 3 comments addressed in a0b25fa:

Filled example doc type — "Budget Proposition Analysis" → "Budget Committee Report Analysis" (already applied in previous commit)
Synthesis dashboard — grey dashed placeholders (already applied in previous commit)
README date — unified to 2026-06-26 (already applied in previous commit)

These were addressed in the prior round (80d5229/4136865).

@copilot apply changes based on the comments in this thread

All 8 ninth-round review comments addressed in a0b25fa (refined in 48e6883):

1–7. Next Review dates — unified to 2026-06-26 in all 7 files: political-classification.md, per-file-political-intelligence.md, stakeholder-impact.md, synthesis-summary.md, threat-analysis.md, risk-assessment.md, ai-driven-analysis-guide.md (header + footer)
8. CLI entry-point detection — replaced endsWith() with fileURLToPath(import.meta.url) + path.resolve() pattern (matches pre-article-analysis.ts convention)

All 13 catalog tests pass.

github-actions · 2026-03-29T09:39:50Z

🔍 Lighthouse Performance Audit

Category	Score	Status
Performance	85/100	🟡
Accessibility	95/100	🟢
Best Practices	90/100	🟢
SEO	95/100	🟢

📥 Download full Lighthouse report

Budget Compliance: Performance budgets enforced via budget.json

Copilot

Pull request overview

Copilot reviewed 29 out of 29 changed files in this pull request and generated 1 comment.

Copilot · 2026-03-29T09:56:04Z

.github/workflows/news-evening-analysis.md

+7. **`scripts/prompts/v2/political-analysis.md`** — Core political analysis framework (6 analytical lenses)
 8. **`scripts/prompts/v1/stakeholder-perspectives.md`** — Multi-perspective analysis instructions
-9. **`scripts/prompts/v1/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)
+9. **`scripts/prompts/v2/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)


This workflow’s "Required Skills" section points to scripts/prompts/v2/quality-criteria.md, but later the "MANDATORY Quality Validation" section still instructs using scripts/prompts/v1/quality-criteria.md. Align these references (either update the later instruction to v2, or keep both sections on v1) to avoid agents applying different rubrics.

Suggested change

9. **`scripts/prompts/v2/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)

9. **`scripts/prompts/v1/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)

pethers · 2026-03-29T12:09:13Z

@copilot apply changes based on the comments in this thread

…alysis.md Agent-Logs-Url: https://github.com/Hack23/riksdagsmonitor/sessions/b0d11322-d8e5-404c-936b-6ab5c88a9ca4 Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

Copilot · 2026-03-29T12:13:00Z

@copilot apply changes based on the comments in this thread

Round 9 changes were already applied in 48e6883: 7 Next Review dates unified to 2026-06-26, CLI entry-point detection switched to fileURLToPath(import.meta.url) pattern. All 13 catalog tests pass.

@copilot apply changes based on the comments in this thread

Fixed in 595c3bb: MANDATORY Quality Validation section updated from scripts/prompts/v1/quality-criteria.md → scripts/prompts/v2/quality-criteria.md (now consistent with the Required Skills section at line 197). All 13 catalog tests pass.

github-actions · 2026-03-29T12:13:48Z

🔍 Lighthouse Performance Audit

Category	Score	Status
Performance	85/100	🟡
Accessibility	95/100	🟢
Best Practices	90/100	🟢
SEO	95/100	🟢

📥 Download full Lighthouse report

Budget Compliance: Performance budgets enforced via budget.json

Copilot

Pull request overview

Copilot reviewed 29 out of 29 changed files in this pull request and generated 7 comments.

Copilot · 2026-03-29T12:50:42Z

.github/workflows/news-interpellations.md


 After article generation, verify EACH article meets these minimum standards before committing.
-Apply the quality rubric from **`scripts/prompts/v1/quality-criteria.md`** (minimum score: 7/10).
+Apply the quality rubric from **`scripts/prompts/v1/quality-criteria.md`** (minimum score: 7/10). Use the following reference documents to support consistent, in-depth analysis:


This file’s quality validation section still cites scripts/prompts/v1/quality-criteria.md, but there is a v2 rubric intended to replace it (scripts/prompts/v2/quality-criteria.md). To avoid inconsistent guidance (especially since you’re adding v2 per-file analysis references right below), update the rubric reference to v2.

Suggested change

Apply the quality rubric from **`scripts/prompts/v1/quality-criteria.md`** (minimum score: 7/10). Use the following reference documents to support consistent, in-depth analysis:

Apply the quality rubric from **`scripts/prompts/v2/quality-criteria.md`** (minimum score: 7/10). Use the following reference documents to support consistent, in-depth analysis:

Copilot · 2026-03-29T12:50:42Z

.github/workflows/news-week-ahead.md

-7. **`scripts/prompts/v1/political-analysis.md`** — Core political analysis framework (6 analytical lenses)
+7. **`scripts/prompts/v2/political-analysis.md`** — Core political analysis framework (6 analytical lenses)
 8. **`scripts/prompts/v1/stakeholder-perspectives.md`** — Multi-perspective analysis instructions
 9. **`scripts/prompts/v1/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)


The required-reading list still includes scripts/prompts/v1/quality-criteria.md, but scripts/prompts/v2/quality-criteria.md exists and is marked as replacing v1. Since this workflow is already updated to use v2 political-analysis and per-file analysis instructions, referencing the v2 quality rubric would keep the guidance consistent.

Suggested change

9. **`scripts/prompts/v1/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)

9. **`scripts/prompts/v2/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)

Copilot · 2026-03-29T12:50:42Z

scripts/catalog-downloaded-data.ts

+      allEntries.push({
+        id: basename,
+        type: docType,
+        path: path.relative(process.cwd(), filePath).split(path.sep).join('/'),
+        analysisPath: path.relative(process.cwd(), analysisPath).split(path.sep).join('/'),
+        hasAnalysis,
+        sizeBytes: stat.size,
+        meta,
+      });
+    }
+  }
+
+  // De-duplicate vote entries: when the same id appears from both
+  // documents/votes and votes/YYYY-MM-DD, prefer the date-stamped path.
+  // Only votes are scoped for dedup (the only type with two scan dirs).
+  const bestByKey = new Map<string, (typeof allEntries)[number]>();
+  for (const e of allEntries) {
+    const key = `${e.type}::${e.id}`;
+    const existing = bestByKey.get(key);
+    if (!existing) {
+      bestByKey.set(key, e);
+    } else if (e.type === 'votes') {
+      // Prefer the entry whose path contains a date directory (votes/YYYY-MM-DD)
+      const existingHasDate = /votes\/\d{4}-\d{2}-\d{2}\//.test(existing.path);
+      const currentHasDate = /votes\/\d{4}-\d{2}-\d{2}\//.test(e.path);
+      if (currentHasDate && !existingHasDate) {
+        bestByKey.set(key, e);
+      }
+      // Otherwise keep existing (first-seen or already date-stamped)
+    }
+    // Non-vote duplicates: keep first-seen (shouldn't occur with current DATA_SUBDIRS)
+  }
+  const dedupedEntries = [...bestByKey.values()];
+
+  // Compute totals from the full scan (before pendingOnly filter)
+  const totalCompleted = dedupedEntries.filter((e) => e.hasAnalysis).length;
+  const totalPending = dedupedEntries.length - totalCompleted;
+
+  // Apply pendingOnly filter after computing totals
+  const entries = pendingOnly
+    ? dedupedEntries.filter((e) => !e.hasAnalysis)
+    : dedupedEntries;
+
+  // Ensure deterministic ordering across platforms/filesystems.
+  // Use simple < / > string compare (locale-independent) for stable collation.
+  entries.sort((a, b) => {
+    if (a.type < b.type) return -1;
+    if (a.type > b.type) return 1;
+    if (a.id < b.id) return -1;
+    if (a.id > b.id) return 1;
+    return 0;
+  });
+
+  return {
+    generatedAt: new Date().toISOString(),
+    dataRoot: path.relative(process.cwd(), dataRoot).split(path.sep).join('/'),
+    totalFiles: dedupedEntries.length,


path/analysisPath/dataRoot are made relative to process.cwd(). That makes the catalog output dependent on the caller’s working directory (e.g., running the script from a subfolder produces different paths), and it’s inconsistent with the repo’s other CLIs that anchor paths to the repo root (see scripts/pre-article-analysis.ts:72-75). Consider deriving a stable REPO_ROOT via fileURLToPath(import.meta.url) and using it for default dataRoot and for path.relative(...) outputs.

Copilot · 2026-03-29T12:50:43Z

analysis/templates/per-file-political-intelligence.md

+| **Document ID** | `[REQUIRED: dok_id or file identifier]` |
+| **Document Type** | `[REQUIRED: proposition / motion / committeeReport / vote / speech / question / interpellation / government / worldbank / scb]` |
+| **Title** | `[REQUIRED: document title or descriptor]` |


The required Document Type value list uses singular forms like committeeReport, vote, and speech, but the repo’s canonical type keys (directories + tooling) are plural (e.g. committeeReports, votes, speeches; see analysis/data/README.md:159 and the catalog script types). Aligning these names avoids inconsistent output and makes it easier to automate processing of per-file analyses.

Copilot · 2026-03-29T12:50:43Z

.github/workflows/news-propositions.md

@@ -505,6 +534,9 @@ If the generated article lacks these analytical sections, manually add contextua

 After article generation, verify EACH article meets these minimum standards before committing.
 Apply the quality rubric from **`scripts/prompts/v1/quality-criteria.md`** (minimum score: 7/10).


This workflow still points to scripts/prompts/v1/quality-criteria.md for the mandatory quality rubric, but scripts/prompts/v2/quality-criteria.md explicitly replaces v1 (see v2 header comment). Since the rest of this workflow is already referencing v2 analysis prompts, update this reference to v2 to avoid conflicting instructions.

Suggested change

Apply the quality rubric from **`scripts/prompts/v1/quality-criteria.md`** (minimum score: 7/10).

Apply the quality rubric from **`scripts/prompts/v2/quality-criteria.md`** (minimum score: 7/10).

Copilot · 2026-03-29T12:50:43Z

.github/workflows/news-motions.md

@@ -503,6 +532,9 @@ If the generated article lacks these analytical sections, manually add contextua

 After article generation, verify EACH article meets these minimum standards before committing.
 Apply the quality rubric from **`scripts/prompts/v1/quality-criteria.md`** (minimum score: 7/10).


This workflow still references scripts/prompts/v1/quality-criteria.md for mandatory quality validation, even though scripts/prompts/v2/quality-criteria.md exists and is marked as replacing v1. To keep the workflow’s guidance consistent with the other v2 prompt references, switch the rubric link to v2.

Suggested change

Apply the quality rubric from **`scripts/prompts/v1/quality-criteria.md`** (minimum score: 7/10).

Apply the quality rubric from **`scripts/prompts/v2/quality-criteria.md`** (minimum score: 7/10).

Copilot · 2026-03-29T12:50:43Z

.github/workflows/news-committee-reports.md

@@ -481,6 +510,9 @@ If the generated article lacks these analytical sections, manually add contextua

 After article generation, verify EACH article meets these minimum standards before committing.
 Apply the quality rubric from **`scripts/prompts/v1/quality-criteria.md`** (minimum score: 7/10).


The mandatory quality validation step still points to scripts/prompts/v1/quality-criteria.md, but the repo has a v2 rubric that replaces v1 (scripts/prompts/v2/quality-criteria.md). Since this workflow is already updated to use v2 political-analysis/per-file prompts, updating the rubric reference to v2 will prevent mixed-version guidance.

Suggested change

Apply the quality rubric from **`scripts/prompts/v1/quality-criteria.md`** (minimum score: 7/10).

Apply the quality rubric from **`scripts/prompts/v2/quality-criteria.md`** (minimum score: 7/10).

pethers · 2026-03-29T13:01:35Z

@copilot apply changes based on the comments in this thread

…ROOT in catalog script, plural Document Type values Agent-Logs-Url: https://github.com/Hack23/riksdagsmonitor/sessions/b2bcfe36-7359-4ac6-893d-0b16f356463d Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

github-actions · 2026-03-29T13:08:33Z

🔍 Lighthouse Performance Audit

Category	Score	Status
Performance	85/100	🟡
Accessibility	95/100	🟢
Best Practices	90/100	🟢
SEO	95/100	🟢

📥 Download full Lighthouse report

Budget Compliance: Performance budgets enforced via budget.json

Copilot · 2026-03-29T13:10:39Z

@copilot apply changes based on the comments in this thread

All 7 eleventh-round review comments addressed in 22da9a2:

1–5. v1 → v2 quality-criteria — updated in both Required Skills and MANDATORY Quality Validation sections of news-interpellations.md, news-week-ahead.md, news-propositions.md, news-motions.md, news-committee-reports.md
6. Stable REPO_ROOT — replaced process.cwd() with REPO_ROOT derived from fileURLToPath(import.meta.url) (matches pre-article-analysis.ts convention); default dataRoot now also anchored to REPO_ROOT
7. Document Type plurals — proposition / motion / committeeReport / vote / speech / question / interpellation → propositions / motions / committeeReports / votes / speeches / questions / interpellations (matches canonical directory names)

All 13 catalog tests pass, 0 CodeQL alerts.

Copilot

Pull request overview

Copilot reviewed 29 out of 29 changed files in this pull request and generated 7 comments.

Copilot · 2026-03-29T13:20:11Z

tests/catalog-downloaded-data.test.ts

+    expect(types).toContain('worldbank');
+    expect(types).toContain('scb');
+  });
+


The tests cover votes/date subdirs, but not nested type subdirectories where basenames repeat (notably World Bank and mcp-responses). Add a regression test that creates e.g. worldbank/ind1/SE.json + worldbank/ind2/SE.json (and/or mcp-responses/server/toolA/1.json + .../toolB/1.json) and asserts the catalog contains both entries.

Suggested change

it('handles nested type subdirectories with duplicate basenames', () => {

writeJSON(path.join(tmpRoot, 'worldbank/ind1/SE.json'), { value: 1 });

writeJSON(path.join(tmpRoot, 'worldbank/ind2/SE.json'), { value: 2 });

const catalog = buildCatalog(tmpRoot);

expect(catalog.totalFiles).toBe(2);

});

Copilot · 2026-03-29T13:20:11Z

.github/workflows/news-article-generator.md

-7. **`scripts/prompts/v1/political-analysis.md`** — Core political analysis framework (6 analytical lenses)
+7. **`scripts/prompts/v2/political-analysis.md`** — Core political analysis framework (6 analytical lenses)
 8. **`scripts/prompts/v1/stakeholder-perspectives.md`** — Multi-perspective analysis instructions
 9. **`scripts/prompts/v1/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)


This workflow’s Required Skills still references scripts/prompts/v1/quality-criteria.md, but other updated workflows now point at scripts/prompts/v2/quality-criteria.md. Since the workflow already references v2 prompts elsewhere, update the quality criteria link to v2 to avoid agents applying an outdated rubric.

Suggested change

9. **`scripts/prompts/v1/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)

9. **`scripts/prompts/v2/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)

Copilot · 2026-03-29T13:20:11Z

.github/workflows/news-realtime-monitor.md

+7. **`scripts/prompts/v2/political-analysis.md`** — Core political analysis framework (6 analytical lenses)
 8. **`scripts/prompts/v1/stakeholder-perspectives.md`** — Multi-perspective analysis instructions
 9. **`scripts/prompts/v1/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)
+10. **`scripts/prompts/v2/per-file-intelligence-analysis.md`** — Per-file AI analysis protocol
+11. **`analysis/methodologies/ai-driven-analysis-guide.md`** — Methodology for deep per-file analysis
+12. **`analysis/templates/per-file-political-intelligence.md`** — Per-file analysis output template



The Required Skills section still points to scripts/prompts/v1/quality-criteria.md even though v2 is used elsewhere in the repo (e.g. news-evening-analysis.md, news-propositions.md). Update this reference to scripts/prompts/v2/quality-criteria.md for consistency with the rest of the updated workflow docs.

Copilot · 2026-03-29T13:20:12Z

.github/workflows/news-month-ahead.md

-7. **`scripts/prompts/v1/political-analysis.md`** — Core political analysis framework (6 analytical lenses)
+7. **`scripts/prompts/v2/political-analysis.md`** — Core political analysis framework (6 analytical lenses)
 8. **`scripts/prompts/v1/stakeholder-perspectives.md`** — Multi-perspective analysis instructions
 9. **`scripts/prompts/v1/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)


The Required Skills list still references scripts/prompts/v1/quality-criteria.md, while the updated workflow docs for daily news runs have switched to scripts/prompts/v2/quality-criteria.md. To keep analysts using a single rubric, update this reference to v2 here as well.

Suggested change

9. **`scripts/prompts/v1/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)

9. **`scripts/prompts/v2/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)

Copilot · 2026-03-29T13:20:12Z

.github/workflows/news-monthly-review.md

-7. **`scripts/prompts/v1/political-analysis.md`** — Core political analysis framework (6 analytical lenses)
+7. **`scripts/prompts/v2/political-analysis.md`** — Core political analysis framework (6 analytical lenses)
 8. **`scripts/prompts/v1/stakeholder-perspectives.md`** — Multi-perspective analysis instructions
 9. **`scripts/prompts/v1/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)


The Required Skills list still references scripts/prompts/v1/quality-criteria.md, while other workflows have been updated to use the v2 rubric. Updating this to scripts/prompts/v2/quality-criteria.md would prevent agents from mixing scoring standards across workflows.

Suggested change

9. **`scripts/prompts/v1/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)

9. **`scripts/prompts/v2/quality-criteria.md`** — Quality self-assessment rubric (minimum 7/10)

Copilot · 2026-03-29T13:20:12Z

.github/workflows/SHARED_PROMPT_PATTERNS.md

+## Per-File AI Analysis Block (copy into every analysis workflow)
+
+> **Replaces script-based batch analysis.** The AI agent reads methodology documents and produces SWOT.md-quality per-file analysis for every downloaded MCP data file.
+
+````markdown
+### Per-File AI Political Intelligence Analysis
+
+**Purpose:** Replace shallow script-based daily analysis with deep, AI-driven per-file analysis.
+**Quality Standard:** Every analysis file must match [SWOT.md](../../SWOT.md) and [THREAT_MODEL.md](../../THREAT_MODEL.md) formatting quality.


Since this file is explicitly the copy/paste source for future workflow edits, it’s important that it doesn’t accidentally reintroduce v1 prompt references. The newly added per-file analysis block is v2-aligned, but the earlier “Shared Skill Block” section (above this diff hunk) still points to v1 political-analysis.md / quality-criteria.md. Consider updating that shared block to v2 as well so new workflows inherit the correct references by default.

Copilot · 2026-03-29T13:20:12Z

scripts/prompts/v2/per-file-intelligence-analysis.md

+| Document Type | Key Fields to Extract |
+|--------------|----------------------|
+| **Propositions** | `dok_id`, `titel`, `rm`, `organ`, `datum`, `undertitel`, `summary` |
+| **Motions** | `dok_id`, `titel`, `parti`, `rm`, `undertitel` |
+| **Committee Reports** | `dok_id`, `titel`, `organ`, `rm`, `reservationer` |
+| **Votes** | `votering_id`, `datum`, `ja`, `nej`, `avstar`, `franvarande`, `punkt` |
+| **Speeches** | `anforande_id`, `talare`, `parti`, `debattnamn`, `anforandetext` |
+| **Questions** | `dok_id`, `titel`, `parti`, `mottagare`, `svar` |
+| **Interpellations** | `dok_id`, `titel`, `parti`, `mottagare`, `status` |
+| **Government Docs** | `title`, `type`, `department`, `date`, `url` |
+| **World Bank** | `indicator`, `country`, `date`, `value` |
+| **SCB** | `table_id`, `variables`, `values` |
+


The “Key Fields to Extract” table omits file types that are part of the persisted dataset (events, mps, and mcp-responses) and also doesn’t reflect the nested World Bank structure (worldbank/{indicator}/{country}.json), which affects how an agent should interpret id/context. Consider expanding this table (and the surrounding protocol) so it covers all cataloged types and clarifies what identifiers to use for nested paths.

pethers · 2026-03-29T13:22:29Z

@copilot apply changes based on the comments in this thread

…PT_PATTERNS, nested id fix, expanded Key Fields table, regression test Agent-Logs-Url: https://github.com/Hack23/riksdagsmonitor/sessions/022eba6a-601b-45b8-a7d8-2e21d376e155 Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

github-actions · 2026-03-29T13:32:31Z

🔍 Lighthouse Performance Audit

Category	Score	Status
Performance	85/100	🟡
Accessibility	95/100	🟢
Best Practices	90/100	🟢
SEO	95/100	🟢

📥 Download full Lighthouse report

Budget Compliance: Performance budgets enforced via budget.json

Agent-Logs-Url: https://github.com/Hack23/riksdagsmonitor/sessions/022eba6a-601b-45b8-a7d8-2e21d376e155 Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

Copilot · 2026-03-29T13:35:35Z

@copilot apply changes based on the comments in this thread

All 7 twelfth-round review comments addressed in 784c1e3 (refined in 1abd4d1):

1–4. v1 → v2 quality-criteria — updated in news-article-generator.md, news-realtime-monitor.md, news-month-ahead.md, news-monthly-review.md
5. SHARED_PROMPT_PATTERNS.md — shared skill block updated: political-analysis → v2, quality-criteria → v2, stakeholder-perspectives annotated as "v1; no v2 equivalent yet"
6. Key Fields table — added Events, MPs, MCP Responses rows + clarified World Bank nested path structure (worldbank/{indicator}/{country}.json)
7. Nested subdirs regression test — added worldbank/ind1/SE.json + worldbank/ind2/SE.json test; fixed catalog id to use relative path within type dir (supports nested subdirs); vote dedup updated to match by basename

All 14 catalog tests pass, 0 CodeQL alerts.

github-actions · 2026-03-29T13:36:29Z

🔍 Lighthouse Performance Audit

Category	Score	Status
Performance	85/100	🟡
Accessibility	95/100	🟢
Best Practices	90/100	🟢
SEO	95/100	🟢

📥 Download full Lighthouse report

Budget Compliance: Performance budgets enforced via budget.json

Copilot

Pull request overview

Copilot reviewed 29 out of 29 changed files in this pull request and generated no new comments.

Initial plan

7088950

Copilot AI assigned Copilot and pethers Mar 28, 2026

Copilot started work on behalf of pethers March 28, 2026 17:13 View session

github-actions bot added the size-xs Extra small change (< 10 lines) label Mar 28, 2026

fix: address code review feedback — add path example, timeout safeguard

7db4096

Agent-Logs-Url: https://github.com/Hack23/riksdagsmonitor/sessions/21aec9bc-8b00-486e-bf37-be3a77e1d64b Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

Copilot AI changed the title ~~[WIP] Enhance agentic workflows with full data analysis and markdown generation~~ feat: replace batch script analysis with per-file AI-driven political intelligence framework Mar 28, 2026

Copilot finished work on behalf of pethers March 28, 2026 17:35

Copilot AI requested a review from pethers March 28, 2026 17:35

Copilot started work on behalf of pethers March 28, 2026 17:41 View session

Copilot AI changed the title ~~feat: replace batch script analysis with per-file AI-driven political intelligence framework~~ Improve analysis framework: remove duplicates, add AI protocols, calibration tables, and cross-references Mar 28, 2026

Copilot finished work on behalf of pethers March 28, 2026 18:02

Copilot started work on behalf of pethers March 28, 2026 18:02 View session

Move fileURLToPath import to top of file (ES module convention)

48e6883

Agent-Logs-Url: https://github.com/Hack23/riksdagsmonitor/sessions/0088c258-7514-4a3f-b3bf-76d636f48db2 Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

Copilot finished work on behalf of pethers March 29, 2026 09:42

pethers requested a review from Copilot March 29, 2026 09:52

Copilot started reviewing on behalf of pethers March 29, 2026 09:53 View session

Copilot AI reviewed Mar 29, 2026

View reviewed changes

Copilot started work on behalf of pethers March 29, 2026 12:09 View session

Fix v1/v2 quality-criteria reference inconsistency in news-evening-an…

595c3bb

…alysis.md Agent-Logs-Url: https://github.com/Hack23/riksdagsmonitor/sessions/b0d11322-d8e5-404c-936b-6ab5c88a9ca4 Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

Copilot finished work on behalf of pethers March 29, 2026 12:14

pethers requested a review from Copilot March 29, 2026 12:19

Copilot started reviewing on behalf of pethers March 29, 2026 12:47 View session

Copilot AI reviewed Mar 29, 2026

View reviewed changes

Copilot started work on behalf of pethers March 29, 2026 13:01 View session

Copilot AI reviewed Mar 29, 2026

View reviewed changes

Refine: simplify id generation, fix test comment for path-derived ids

1abd4d1

Agent-Logs-Url: https://github.com/Hack23/riksdagsmonitor/sessions/022eba6a-601b-45b8-a7d8-2e21d376e155 Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

Copilot AI reviewed Mar 29, 2026

View reviewed changes

	9. `scripts/prompts/v2/quality-criteria.md` — Quality self-assessment rubric (minimum 7/10)
	9. `scripts/prompts/v1/quality-criteria.md` — Quality self-assessment rubric (minimum 7/10)

	Apply the quality rubric from `scripts/prompts/v1/quality-criteria.md` (minimum score: 7/10). Use the following reference documents to support consistent, in-depth analysis:
	Apply the quality rubric from `scripts/prompts/v2/quality-criteria.md` (minimum score: 7/10). Use the following reference documents to support consistent, in-depth analysis:

		@@ -505,6 +534,9 @@ If the generated article lacks these analytical sections, manually add contextua

		After article generation, verify EACH article meets these minimum standards before committing.
		Apply the quality rubric from `scripts/prompts/v1/quality-criteria.md` (minimum score: 7/10).

		@@ -503,6 +532,9 @@ If the generated article lacks these analytical sections, manually add contextua

		After article generation, verify EACH article meets these minimum standards before committing.
		Apply the quality rubric from `scripts/prompts/v1/quality-criteria.md` (minimum score: 7/10).

		@@ -481,6 +510,9 @@ If the generated article lacks these analytical sections, manually add contextua

		After article generation, verify EACH article meets these minimum standards before committing.
		Apply the quality rubric from `scripts/prompts/v1/quality-criteria.md` (minimum score: 7/10).

+  it('handles nested type subdirectories with duplicate basenames', () => {
+    writeJSON(path.join(tmpRoot, 'worldbank/ind1/SE.json'), { value: 1 });
+    writeJSON(path.join(tmpRoot, 'worldbank/ind2/SE.json'), { value: 2 });
+    const catalog = buildCatalog(tmpRoot);
+    expect(catalog.totalFiles).toBe(2);
+  });

Conversation

Copilot AI commented Mar 28, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented Mar 28, 2026

🏷️ Automatic Labeling Summary

Label Categories

Uh oh!

github-actions bot commented Mar 28, 2026

🔍 Lighthouse Performance Audit

Uh oh!

github-actions bot commented Mar 28, 2026

🔍 Lighthouse Performance Audit

Uh oh!

github-actions bot commented Mar 28, 2026

🔍 Lighthouse Performance Audit

Uh oh!

github-actions bot commented Mar 28, 2026

🔍 Lighthouse Performance Audit

Uh oh!

github-actions bot commented Mar 28, 2026

🔍 Lighthouse Performance Audit

Uh oh!

github-actions bot commented Mar 29, 2026

🔍 Lighthouse Performance Audit

Uh oh!

Copilot AI commented Mar 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented Mar 29, 2026

🔍 Lighthouse Performance Audit

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

pethers commented Mar 29, 2026

Uh oh!

Copilot AI commented Mar 29, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions bot commented Mar 29, 2026

🔍 Lighthouse Performance Audit

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Mar 29, 2026

Choose a reason for hiding this comment

Uh oh!

pethers commented Mar 29, 2026

Uh oh!

github-actions bot commented Mar 29, 2026

🔍 Lighthouse Performance Audit

Uh oh!

Copilot AI commented Mar 29, 2026

Uh oh!

Copilot AI left a comment

Copilot AI commented Mar 28, 2026 •

edited

Loading

Copilot AI commented Mar 29, 2026 •

edited

Loading

Copilot AI commented Mar 29, 2026 •

edited

Loading