feat(committee-reports): implement search_voteringar, search_anforanden, get_propositioner enrichment by Copilot · Pull Request #608 · Hack23/riksdagsmonitor

Copilot · 2026-02-26T21:01:11Z

Original prompt

This section details on the original issue you should resolve

<issue_title>Improve Committee Reports: Add voting patterns, debate context, and proposition linkage</issue_title>
<issue_description>## 📋 Issue Type
Enhancement — Implement missing MCP tools for committee reports: voting patterns, debate context, and proposition linkage

🎯 Objective

Implement the three TODO MCP tools in scripts/news-types/committee-reports.ts: search_voteringar (committee voting patterns), search_anforanden (committee member statements and debate), and get_propositioner (linkage to originating government proposals). These tools will transform committee report articles from basic report listings into comprehensive legislative intelligence.

📊 Current State

File: scripts/news-types/committee-reports.ts (425 lines)
Implemented tools: get_betankanden ✅ (only tool)
Missing tools: search_voteringar ❌, search_anforanden ❌, get_propositioner ❌
REQUIRED_TOOLS: Only 1 tool (get_betankanden) — reduced from original 4 to avoid validation failures
Impact: Committee report articles list betänkanden metadata without voting results, debate highlights, or connection to originating government proposals

🚀 Desired State

Voting pattern analysis shows how each party voted on committee recommendations
Debate context captures key committee member arguments and dissenting views
Proposition linkage traces committee reports back to originating government bills
Minority report analysis identifies party splits within committees
Committee power mapping (which committees are most active/influential)
Cross-committee pattern detection (related reports from different committees)
"Legislative Impact" assessment for each betänkande

🔧 Implementation Approach

1. Add `search_voteringar` integration (~80 lines)

// Fetch voting records for committee report decisions
const votes = await mcpClient.call('search_voteringar', {
  bet: betankande.beteckning,
  rm: currentRiksmote,
  limit: 20
});

Parse party-level voting on each committee recommendation
Identify unanimous vs contentious decisions
Detect coalition vs opposition voting patterns
Calculate committee consensus scores

2. Add `search_anforanden` integration (~70 lines)

// Find committee debate speeches
const debates = await mcpClient.call('search_anforanden', {
  debattnamn: betankande.debattnamn || betankande.titel,
  rm: currentRiksmote,
  limit: 15
});

Extract committee chair statements
Capture minority report arguments
Identify key points of contention
Quote notable committee member interventions

3. Add `get_propositioner` integration (~50 lines)

// Link committee reports to originating government proposals
const propositioner = await mcpClient.call('get_propositioner', {
  rm: currentRiksmote,
  limit: 20
});
// Cross-reference by subject matter
const relatedPropositions = matchPropositionsToReports(propositioner, betankanden);

Trace each betänkande back to its originating proposition
Show how committee modified or endorsed government proposals
Identify where committee rejected or amended government bills

4. Update REQUIRED_TOOLS

Restore full 4-tool list: get_betankanden, search_voteringar, search_anforanden, get_propositioner

5. Enhanced article templates

Add "Voting Results" section with party breakdown per betänkande
Add "Committee Debate" highlights section
Add "Government Bill Linkage" tracing section
Add "Minority Reports" analysis where applicable
Add "Committee Spotlight" for most active/influential committees
Update all 14 language templates

🤖 Recommended Agent

code-quality-engineer — TypeScript implementation matching existing MCP integration patterns

✅ Acceptance Criteria

📚 References

File: scripts/news-types/committee-reports.ts:78-81 (TODO comments)
File: scripts/news-types/committee-reports.ts:187-196 (REQUIRED_TOOLS update comment)
MCP Tools: search_voteringar, search_anforanden, get_propositioner
Content generators: scripts/data-transformers/content-generators.ts
Policy analysis:...

Fixes Improve Committee Reports: Add voting patterns, debate context, and proposition linkage #602

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.

…committee-reports Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

github-actions · 2026-02-27T02:15:07Z

🔍 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

Implements the previously-missing MCP enrichment for committee report articles so they include voting records, debate/speech context, and proposition linkage, and updates the data pipeline/types/tests accordingly.

Changes:

Restores REQUIRED_TOOLS for committee reports to the full 4-tool set and records MCP calls for all tools.
Fetches votes, speeches, and propositions in parallel and passes them into article content generation.
Extends ArticleContentData and committee content generation with optional enrichment sections; updates and expands unit tests.

Reviewed changes

Copilot reviewed 4 out of 4 changed files in this pull request and generated 3 comments.

File	Description
`scripts/news-types/committee-reports.ts`	Restores 4-tool requirement and adds parallel enrichment (votes/speeches/propositions) into committee report generation.
`scripts/data-transformers/types.ts`	Extends `ArticleContentData` with optional `votes` and `speeches` fields.
`scripts/data-transformers/content-generators.ts`	Adds optional per-language sections for voting results, debate context, and proposition linkage in committee reports content.
`tests/news-types/committee-reports.test.ts`	Updates mocks and assertions to validate the expanded tool usage and cross-reference tracking.

Copilot · 2026-02-27T02:17:22Z

scripts/news-types/committee-reports.ts

+    // Step 2: Enrich with voting patterns, speeches, and propositions (non-fatal)
+    console.log('  🔄 Fetching voting patterns, speeches, and propositions...');
+    const currentRm = '2025/26';
+    const [votes, speeches, propositions] = await Promise.all([
+      Promise.resolve()
+        .then(() => client.fetchVotingRecords({ rm: currentRm, limit: 20 }) as Promise<unknown[]>)
+        .catch((err: unknown) => { console.error('  ⚠️ Failed to fetch voting records:', (err as Error)?.message ?? String(err)); return [] as unknown[]; }),
+      Promise.resolve()
+        .then(() => client.searchSpeeches({ rm: currentRm, limit: 15 }) as Promise<unknown[]>)
+        .catch((err: unknown) => { console.error('  ⚠️ Failed to fetch speeches:', (err as Error)?.message ?? String(err)); return [] as unknown[]; }),
+      Promise.resolve()
+        .then(() => client.fetchPropositions(20) as Promise<unknown[]>)
+        .catch((err: unknown) => { console.error('  ⚠️ Failed to fetch propositions:', (err as Error)?.message ?? String(err)); return [] as unknown[]; }),


currentRm is hard-coded to '2025/26', while the report fetch (fetchCommitteeReports(limit)) is not scoped to the same rm. This can produce enrichment data (votes/speeches/propositions) from a different parliamentary session than the reports, and will silently become stale when the session changes. Consider deriving rm from the fetched report data (if available), or passing the same rm consistently into all three enrichment calls (and fetchCommitteeReports) / making it an option.

Fixed in ab8bff6. currentRm is now derived from the first fetched report's rm field when available, falling back to a date-based calculation (month ≥ 8/September → current year, else previous year) matching the parliamentary calendar. This ensures all three enrichment calls use the same session as the reports themselves.

Copilot · 2026-02-27T02:17:22Z

scripts/news-types/committee-reports.ts

+        .then(() => client.searchSpeeches({ rm: currentRm, limit: 15 }) as Promise<unknown[]>)
+        .catch((err: unknown) => { console.error('  ⚠️ Failed to fetch speeches:', (err as Error)?.message ?? String(err)); return [] as unknown[]; }),
+      Promise.resolve()
+        .then(() => client.fetchPropositions(20) as Promise<unknown[]>)


fetchPropositions supports an rm parameter (limit, rm), but the call here omits it even though currentRm is defined and used for votes/speeches. This makes proposition enrichment potentially inconsistent with the rest of the enrichment set; pass currentRm (or remove currentRm entirely and rely on server defaults consistently).

Suggested change

.then(() => client.fetchPropositions(20) as Promise<unknown[]>)

.then(() => client.fetchPropositions(20, currentRm) as Promise<unknown[]>)

Fixed in ab8bff6 — currentRm is now passed as the second argument to fetchPropositions(20, currentRm), keeping all three enrichment calls consistent.

Copilot · 2026-02-27T02:17:22Z

scripts/data-transformers/content-generators.ts

+  // ── Optional: Voting Results section ─────────────────────────────────────
+  const votes = (data.votes ?? []) as unknown[];
+  if (votes.length > 0) {
+    const votingSectionHeaders: Record<string, string> = {
+      sv: 'Röstningsresultat', da: 'Afstemningsresultater', no: 'Voteringsresultater',
+      fi: 'Äänestystulokset', de: 'Abstimmungsergebnisse', fr: 'Résultats du vote',
+      es: 'Resultados de la votación', nl: 'Stemresultaten', ar: 'نتائج التصويت',
+      he: 'תוצאות ההצבעה', ja: '投票結果', ko: '투표 결과', zh: '投票结果',
+    };
+    const votingCountTemplates: Record<string, (n: number) => string> = {
+      sv: (n) => `${n} röstningsprotokoll visar hur partierna röstade i utskottsbeslut denna period.`,
+      da: (n) => `${n} afstemningsprotokoller viser, hvordan partierne stemte om udvalgets beslutninger.`,
+      no: (n) => `${n} voteringsprotokoll viser hvordan partiene stemte i komitévedtak.`,
+      fi: (n) => `${n} äänestysrekisteriä osoittaa, miten puolueet äänestivät valiokunnan päätöksistä.`,
+      de: (n) => `${n} Abstimmungsrekorde zeigen, wie die Parteien über Ausschussbeschlüsse abstimmten.`,
+      fr: (n) => `${n} procès-verbaux de vote montrent comment les partis ont voté sur les décisions de commission.`,
+      es: (n) => `${n} registros de votación muestran cómo votaron los partidos en las decisiones de la comisión.`,
+      nl: (n) => `${n} stemregisters tonen hoe partijen stemden over commissiebeslissingen.`,
+      ar: (n) => `${n} سجلات التصويت تظهر كيف صوتت الأحزاب على قرارات اللجنة.`,
+      he: (n) => `${n} פרוטוקולי הצבעה מציגים כיצד הצביעו המפלגות על החלטות הוועדה.`,
+      ja: (n) => `${n}件の投票記録が、委員会決定に対する各党の投票方法を示しています。`,
+      ko: (n) => `${n}건의 투표 기록이 위원회 결정에 대한 각 정당의 투표 방식을 보여줍니다.`,
+      zh: (n) => `${n}条投票记录显示各党派对委员会决定的投票情况。`,
+    };
+    const votingHeader = votingSectionHeaders[lang as string] ?? 'Voting Results';
+    const votingCountFn = votingCountTemplates[lang as string];
+    const votingCountText = votingCountFn
+      ? votingCountFn(votes.length)
+      : `${votes.length} voting records show how parties voted on committee decisions this period.`;
+    content += `\n    <h2>${escapeHtml(votingHeader)}</h2>\n`;
+    content += `    <p>${escapeHtml(votingCountText)}</p>\n`;
+  }
+
+  // ── Optional: Committee Debate section ───────────────────────────────────
+  const speeches = (data.speeches ?? []) as unknown[];
+  if (speeches.length > 0) {
+    const debateSectionHeaders: Record<string, string> = {
+      sv: 'Utskottsdebatt', da: 'Udvalgets debat', no: 'Komitédebatt',
+      fi: 'Valiokunnan keskustelu', de: 'Ausschussdebatte', fr: 'Débat en commission',
+      es: 'Debate en comisión', nl: 'Commissiedebat', ar: 'نقاش اللجنة',
+      he: 'דיון בוועדה', ja: '委員会討論', ko: '위원회 토론', zh: '委员会讨论',
+    };
+    const debateCountTemplates: Record<string, (n: number) => string> = {
+      sv: (n) => `${n} anföranden i kammaren belyser de viktigaste argumenten och partipositionerna i dessa frågor.`,
+      da: (n) => `${n} parlamentariske taler belyser nøgleargumenter og partipositioner.`,
+      no: (n) => `${n} parlamentariske innlegg belyser nøkkelargumenter og partiposisjoner.`,
+      fi: (n) => `${n} parlamentaarista puheenvuoroa valaisee keskeisiä argumentteja ja puolueiden kantoja.`,
+      de: (n) => `${n} parlamentarische Reden beleuchten Hauptargumente und Parteipositionen.`,
+      fr: (n) => `${n} discours parlementaires éclairent les arguments clés et les positions des partis.`,
+      es: (n) => `${n} discursos parlamentarios iluminan los principales argumentos y posiciones de los partidos.`,
+      nl: (n) => `${n} parlementaire toespraken belichten de belangrijkste argumenten en partijposities.`,
+      ar: (n) => `${n} خطاب برلماني يسلط الضوء على الحجج الرئيسية ومواقف الأحزاب.`,
+      he: (n) => `${n} נאומים פרלמנטריים מאירים טיעונים מרכזיים ועמדות מפלגות.`,
+      ja: (n) => `${n}件の議会演説が主要な論点と各党の立場を明らかにしています。`,
+      ko: (n) => `${n}건의 의회 연설이 주요 논점과 각 정당의 입장을 보여줍니다.`,
+      zh: (n) => `${n}篇议会演讲揭示了主要论点和各党派立场。`,
+    };
+    const debateHeader = debateSectionHeaders[lang as string] ?? 'Committee Debate';
+    const debateCountFn = debateCountTemplates[lang as string];
+    const debateCountText = debateCountFn
+      ? debateCountFn(speeches.length)
+      : `${speeches.length} parliamentary speeches highlight key arguments and party positions on these issues.`;
+    content += `\n    <h2>${escapeHtml(debateHeader)}</h2>\n`;
+    content += `    <p>${escapeHtml(debateCountText)}</p>\n`;
+  }
+


New optional sections (Voting Results / Committee Debate / Government Bill Linkage) are rendered based on votes, speeches, and propositions, but there are no unit tests asserting that these sections appear when non-empty and are omitted when empty. Add focused tests (e.g., in tests/data-transformers.test.ts) that pass in minimal non-empty arrays and assert the expected headings/text are present/absent.

Added in ab8bff6. Eight new tests in tests/data-transformers.test.ts assert that each optional section (Voting Results, Committee Debate, Government Bill Linkage) is rendered when its respective array is non-empty and omitted when empty, plus Swedish heading variants for sv locale.

pethers · 2026-02-27T06:41:09Z

@copilot apply changes based on the comments in this thread

…nrichment section tests Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

Copilot · 2026-02-27T06:45:50Z

@copilot apply changes based on the comments in this thread

Done in ab8bff6. All three review comments addressed: dynamic riksmöte derivation, rm passed to fetchPropositions, and 8 new unit tests for the optional content sections.

Copilot

Pull request overview

Copilot reviewed 5 out of 5 changed files in this pull request and generated 2 comments.

Copilot · 2026-02-27T07:02:02Z

tests/news-types/committee-reports.test.ts

  fetchCommitteeReports: Mock<(limit: number) => Promise<CommitteeReport[]>>;
+  fetchVotingRecords: Mock<(filters: object) => Promise<unknown[]>>;
+  searchSpeeches: Mock<(params: object) => Promise<unknown[]>>;
+  fetchPropositions: Mock<(limit: number) => Promise<unknown[]>>;


The mock signature for fetchPropositions is incomplete. The actual method signature is fetchPropositions(limit = 10, rm: string | null = null), which takes two parameters. The mock should reflect this to accurately test that the rm parameter is being passed correctly.

Suggested change

fetchPropositions: Mock<(limit: number) => Promise<unknown[]>>;

fetchPropositions: Mock<(limit?: number, rm?: string | null) => Promise<unknown[]>>;

Fixed in ffd42c9 — mock signature updated to Mock<(limit?: number, rm?: string | null) => Promise<unknown[]>>.

Copilot · 2026-02-27T07:02:03Z

tests/news-types/committee-reports.test.ts

+    it('should call search_voteringar for voting patterns', async () => {
+      const result = await committeeReportsModule.generateCommitteeReports({
+        languages: ['en']
+      });
+
+      expect(mockClientInstance.fetchVotingRecords).toHaveBeenCalled();
+      expect(result.mcpCalls!.some((call: MCPCallRecord) => call.tool === 'search_voteringar')).toBe(true);
+    });
+
+    it('should call search_anforanden for debate context', async () => {
+      const result = await committeeReportsModule.generateCommitteeReports({
+        languages: ['en']
+      });
+
+      expect(mockClientInstance.searchSpeeches).toHaveBeenCalled();
+      expect(result.mcpCalls!.some((call: MCPCallRecord) => call.tool === 'search_anforanden')).toBe(true);
+    });
+
+    it('should call get_propositioner for bill linkage', async () => {
+      const result = await committeeReportsModule.generateCommitteeReports({
+        languages: ['en']
+      });
+
+      expect(mockClientInstance.fetchPropositions).toHaveBeenCalled();
+      expect(result.mcpCalls!.some((call: MCPCallRecord) => call.tool === 'get_propositioner')).toBe(true);
+    });


While the tests verify that enrichment methods are called, they don't verify that the rm parameter is correctly passed to fetchVotingRecords, searchSpeeches, and fetchPropositions. Consider adding assertions like expect(mockClientInstance.fetchPropositions).toHaveBeenCalledWith(20, '2024/25') to verify the riksmöte session parameter is consistently passed to all enrichment calls.

Fixed in ffd42c9. Assertions now verify the rm parameter is passed consistently to all three enrichment calls:

fetchVotingRecords — expect.objectContaining({ rm: '2024/25', limit: 20 })

searchSpeeches — expect.objectContaining({ rm: '2024/25', limit: 15 })

fetchPropositions — toHaveBeenCalledWith(20, '2024/25')

The '2024/25' value is derived from the mock report data's rm field, matching the actual runtime behaviour.

pethers · 2026-02-27T07:13:44Z

@copilot apply changes based on the comments in this thread

…ram in enrichment calls Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

Copilot · 2026-02-27T07:18:53Z

@copilot apply changes based on the comments in this thread

Done in ffd42c9 — both review comments addressed.

Signed-off-by: James Pether Sörling <pethers@users.noreply.github.com>

github-actions · 2026-02-27T08:28:54Z

🔍 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 5 out of 5 changed files in this pull request and generated 4 comments.

Copilot · 2026-02-27T08:30:44Z

scripts/news-types/committee-reports.ts

+    console.log('  🔄 Fetching voting patterns, speeches, and propositions...');
+    // Derive riksmöte from the first report's rm field, or calculate from current date.
+    // Parliamentary year starts in September; e.g. Sep 2025–Aug 2026 → '2025/26'.
+    const firstReportRm = (reports[0] as Record<string, unknown>)?.['rm'] as string | undefined;


firstReportRm is read via a forced cast (as string | undefined). Since rm is not part of the RawDocument type here, this can silently pass a non-string (or an invalid session format) into the enrichment calls. Prefer extracting rm as unknown and validating typeof rm === 'string' (and optionally a simple YYYY/YY format check) before using it; otherwise fall back to the date-derived value.

Suggested change

const firstReportRm = (reports[0] as Record<string, unknown>)?.['rm'] as string | undefined;

const firstReportRaw = (reports[0] as Record<string, unknown>)?.['rm'];

let firstReportRm: string | undefined;

if (typeof firstReportRaw === 'string') {

// Basic riksmöte format validation: 'YYYY/YY', e.g. '2025/26'

const rmPattern = /^\d{4}\/\d{2}$/;

if (rmPattern.test(firstReportRaw)) {

firstReportRm = firstReportRaw;

}

}

Fixed in d39d175. firstReportRm is now validated as a string matching /^\d{4}\/\d{2}$/ before use; invalid or non-string values fall through to the date-based calculation.

Copilot · 2026-02-27T08:30:45Z

scripts/news-types/committee-reports.ts

+      const content: string = generateArticleContent(
+        { reports, votes, speeches, propositions: propositions as RawDocument[] },
+        'committee-reports',


propositions is unknown[] from fetchPropositions, but it’s passed into generateArticleContent as propositions: propositions as RawDocument[]. If the MCP response ever contains non-object entries, downstream rendering (which reads prop.titel/title/dokumentnamn) will throw at runtime. Consider filtering/normalizing propositions to plain objects (or to RawDocument-shaped records) before passing it into the content generator.

Fixed in d39d175. propositions is now filtered to only plain objects (safePropositions) before being passed to the content generator, eliminating the unsafe cast.

Copilot · 2026-02-27T08:30:45Z

scripts/data-transformers/content-generators.ts

+    propositions.slice(0, 3).forEach(prop => { // display up to 3 linked propositions
+      const propTitle = escapeHtml(prop.titel || prop.title || prop.dokumentnamn || '');
+      if (propTitle) {
+        content += `    <p>→ ${propTitle}</p>\n`;
+      }
+    });


propositions is treated as RawDocument[], but the loop accesses prop.titel/prop.title/prop.dokumentnamn directly. If any entry is not a non-null object, this will throw. Consider guarding inside the loop (e.g., skip non-object entries) or pre-filtering propositions to only objects with string fields before rendering.

Fixed in d39d175. The loop now skips entries where typeof prop !== 'object' || prop === null before accessing title fields.

Copilot · 2026-02-27T08:30:45Z

scripts/news-types/committee-reports.ts

      const metadata = generateMetadata({ reports }, 'committee-reports', lang);
      const readTime: string = calculateReadTime(content);
-      const sources: string[] = generateSources(['get_betankanden']);
+      const sources: string[] = generateSources(['get_betankanden', 'search_voteringar', 'search_anforanden', 'get_propositioner']);


generateSources() currently only adds specific source labels for a subset of tool names (e.g. get_betankanden, get_propositioner, search_dokument), and does not handle search_voteringar or search_anforanden. Passing those tool names here is therefore misleading (no additional sources will be listed). Either extend generateSources to map these tools to appropriate source labels, or omit them from this call to avoid implying they’re reflected in the sources list.

Suggested change

const sources: string[] = generateSources(['get_betankanden', 'search_voteringar', 'search_anforanden', 'get_propositioner']);

const sources: string[] = generateSources(['get_betankanden', 'get_propositioner']);

Fixed in d39d175. Extended generateSources in metadata.ts to map search_voteringar → 'Riksdagen Voting Records' and search_anforanden → 'Riksdagen Speeches', so both tools now produce meaningful source labels.

Signed-off-by: James Pether Sörling <pethers@users.noreply.github.com>

pethers · 2026-02-27T10:16:37Z

@copilot apply changes based on the comments in this thread

github-actions · 2026-02-27T10:17:49Z

🔍 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

…end generateSources Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

Copilot · 2026-02-27T10:21:55Z

@copilot apply changes based on the comments in this thread

Done in d39d175 — all four review comments addressed.

github-actions · 2026-02-27T10:27:29Z

🔍 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 6 out of 6 changed files in this pull request and generated no new comments.

Copilot AI assigned Copilot and pethers Feb 26, 2026

Copilot started work on behalf of pethers February 26, 2026 21:01 View session

Copilot AI mentioned this pull request Feb 26, 2026

Create 8 issues to improve/extend all article types with MCP tool integrations #594

Draft

feat: add search_voteringar, search_anforanden, get_propositioner to …

0e2aebc

…committee-reports Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

Copilot AI changed the title ~~[WIP] Enhance committee reports with missing MCP tools~~ feat(committee-reports): implement search_voteringar, search_anforanden, get_propositioner enrichment Feb 26, 2026

Copilot finished work on behalf of pethers February 26, 2026 21:13

pethers requested a review from Copilot February 27, 2026 02:13

github-actions bot added testing Test coverage refactor Code refactoring size-m Medium change (50-250 lines) labels Feb 27, 2026

Copilot started reviewing on behalf of pethers February 27, 2026 02:14 View session

Copilot AI reviewed Feb 27, 2026

View reviewed changes

Copilot started work on behalf of pethers February 27, 2026 06:41 View session

fix: derive riksmöte dynamically, pass rm to fetchPropositions, add e…

ab8bff6

…nrichment section tests Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

Copilot finished work on behalf of pethers February 27, 2026 06:47

pethers requested a review from Copilot February 27, 2026 06:57

github-actions bot added the size-l Large change (250-1000 lines) label Feb 27, 2026

Copilot started reviewing on behalf of pethers February 27, 2026 06:58 View session

Copilot AI reviewed Feb 27, 2026

View reviewed changes

Copilot started work on behalf of pethers February 27, 2026 07:15 View session

fix(tests): correct fetchPropositions mock signature and assert rm pa…

ffd42c9

…ram in enrichment calls Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

Copilot finished work on behalf of pethers February 27, 2026 07:19

Merge branch 'main' into copilot/add-missing-mcp-tools-committee-reports

8aaee62

Signed-off-by: James Pether Sörling <pethers@users.noreply.github.com>

pethers requested a review from Copilot February 27, 2026 08:27

Copilot started reviewing on behalf of pethers February 27, 2026 08:28 View session

Copilot AI reviewed Feb 27, 2026

View reviewed changes

Merge branch 'main' into copilot/add-missing-mcp-tools-committee-reports

52b1289

Signed-off-by: James Pether Sörling <pethers@users.noreply.github.com>

Copilot started work on behalf of pethers February 27, 2026 10:16 View session

fix: validate rm format, normalize propositions, guard prop loop, ext…

d39d175

…end generateSources Co-authored-by: pethers <1726836+pethers@users.noreply.github.com>

Copilot finished work on behalf of pethers February 27, 2026 10:23

pethers requested a review from Copilot February 27, 2026 10:26

Copilot started reviewing on behalf of pethers February 27, 2026 10:26 View session

Copilot AI reviewed Feb 27, 2026

View reviewed changes

pethers marked this pull request as ready for review February 27, 2026 10:45

pethers merged commit 7c587b6 into main Feb 27, 2026
20 checks passed

pethers deleted the copilot/add-missing-mcp-tools-committee-reports branch February 27, 2026 10:45

	.then(() => client.fetchPropositions(20) as Promise<unknown[]>)
	.then(() => client.fetchPropositions(20, currentRm) as Promise<unknown[]>)

	fetchPropositions: Mock<(limit: number) => Promise<unknown[]>>;
	fetchPropositions: Mock<(limit?: number, rm?: string \| null) => Promise<unknown[]>>;

-    const firstReportRm = (reports[0] as Record<string, unknown>)?.['rm'] as string | undefined;
+    const firstReportRaw = (reports[0] as Record<string, unknown>)?.['rm'];
+    let firstReportRm: string | undefined;
+    if (typeof firstReportRaw === 'string') {
+      // Basic riksmöte format validation: 'YYYY/YY', e.g. '2025/26'
+      const rmPattern = /^\d{4}\/\d{2}$/;
+      if (rmPattern.test(firstReportRaw)) {
+        firstReportRm = firstReportRaw;
+      }
+    }

	const sources: string[] = generateSources(['get_betankanden', 'search_voteringar', 'search_anforanden', 'get_propositioner']);
	const sources: string[] = generateSources(['get_betankanden', 'get_propositioner']);

Conversation

Copilot AI commented Feb 26, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

🎯 Objective

📊 Current State

🚀 Desired State

🔧 Implementation Approach

1. Add search_voteringar integration (~80 lines)

2. Add search_anforanden integration (~70 lines)

3. Add get_propositioner integration (~50 lines)

4. Update REQUIRED_TOOLS

5. Enhanced article templates

🤖 Recommended Agent

✅ Acceptance Criteria

📚 References

Uh oh!

github-actions bot commented Feb 27, 2026

🔍 Lighthouse Performance Audit

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Reviewed changes

Uh oh!

Copilot AI Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

pethers commented Feb 27, 2026

Uh oh!

Copilot AI commented Feb 27, 2026

Uh oh!

Copilot AI left a comment

Choose a reason for hiding this comment

Pull request overview

Uh oh!

Copilot AI Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

pethers commented Feb 27, 2026

Uh oh!

Copilot AI commented Feb 27, 2026

Uh oh!

github-actions bot commented Feb 27, 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 Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 27, 2026

Choose a reason for hiding this comment

Uh oh!

Copilot AI Feb 27, 2026

Copilot AI commented Feb 26, 2026 •

edited

Loading

1. Add `search_voteringar` integration (~80 lines)

2. Add `search_anforanden` integration (~70 lines)

3. Add `get_propositioner` integration (~50 lines)