Harden news workflow pass‑2/re-run contracts and rebalance schedule/model budget for reliable 60‑min runs by Copilot · Pull Request #2495 · Hack23/riksdagsmonitor

Copilot · 2026-05-14T23:24:19Z

This change addresses recurring 2026-05-14 cohort failures where workflows silently shipped incomplete Pass‑2 output, showed inconsistent re-run metadata, and missed midday chamber updates in realtime monitoring. The update tightens shared contracts and adjusts workflow timing/model choices to improve deterministic delivery within the 60-minute envelope.

Pass‑2 gate hardening (fail-loud)
- Added explicit shared requirement that methodology-reflection.md declares:
  - Pass-2 status: executed in full
- Gate now rejects statuses indicating non-execution/deferral/partial completion.
Canonical re-run schema normalization (all news workflows)
- Standardized improvement-mode ## Re-run log contract in shared prompt/template surfaces.
- Enforced canonical fields in gate text contract:
  - run_id, attempt, new dok_ids, artifacts extended, flags closed, vintage refresh.
Realtime monitor coverage window adjustment
- Moved weekday afternoon run to 15:30 UTC to better capture same-day 13:00–15:00 UTC interpellation/chamber updates in the primary run.
Budget/throughput tuning for long-horizon workflows
- Switched news-election-cycle and news-year-ahead engine model to claude-sonnet-4.6 to reduce runtime pressure while preserving two-pass workflow semantics.
- Clarified election-cycle scope policy: anchor omission cannot be justified by time budget.
Cycle-rollover policy clarification
- Added explicit anchor-coverage rule in rollover module: when cycle_anchor=both, skipping an anchor requires formal rollover-window reason, not runtime exhaustion wording.
Contract-level regression coverage
- Added focused test to lock new invariants:
  - realtime cron window
  - model selection for election-cycle/year-ahead
  - pass-2/re-run schema contract presence in gate/template.

## Re-run log

- **Re-run**: 2026-05-14 15:02 UTC · workflow=news-realtime-monitor · run_id=25864884905 · attempt=1
  - new dok_ids: 3
  - artifacts extended: synthesis-summary.md, forward-indicators.md, methodology-reflection.md
  - flags closed: 2
  - vintage refresh: yes

github-actions · 2026-05-14T23:25:18Z

🏷️ 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 · 2026-05-14T23:26:12Z

🔍 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

… window Agent-Logs-Url: https://github.com/Hack23/riksdagsmonitor/sessions/e51960bd-f4d0-48d4-9b62-9767d7a0283a Co-authored-by: pethers <[email protected]>

github-actions · 2026-05-15T00:00:30Z

🔍 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

github-actions · 2026-05-15T06:15:24Z

🔍 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

This PR strengthens the agentic news workflow reliability contract by hardening Pass‑2/improvement-mode gating, standardizing re-run metadata requirements, and rebalancing schedule/model settings to better fit the 60‑minute runtime envelope.

Changes:

Tightens the analysis gate to require an explicit Pass‑2 completion declaration and (in improvement-mode) a canonical ## Re-run log schema.
Adjusts operational knobs: moves realtime monitor weekday afternoon cron to 15:30 UTC and switches long-horizon workflows to claude-sonnet-4.6.
Adds a focused Vitest contract test to lock the new schedule/model/contract invariants.

Reviewed changes

Copilot reviewed 23 out of 23 changed files in this pull request and generated 18 comments.

Show a summary per file

File	Description
tests/news-workflow-pass2-budget-contract.test.ts	Adds contract assertions for cron timing, model selection, and Pass‑2/re-run schema presence.
analysis/templates/methodology-reflection.md	Updates the methodology reflection template with Pass‑2 status and re-run log sections.
.github/workflows/news-realtime-monitor.md	Moves the weekday afternoon schedule to 15:30 UTC for better same-day chamber coverage.
.github/workflows/news-election-cycle.md	Switches engine model to Sonnet and clarifies anchor-coverage scope policy.
.github/workflows/news-year-ahead.md	Switches engine model to Sonnet for throughput/budget tuning.
.github/prompts/04-analysis-pipeline.md	Adds an explicit Pass‑2 declaration step and parallelisation guidance.
.github/prompts/05-analysis-gate.md	Hardens the gate to require explicit Pass‑2 declaration and improvement-mode rerun schema fields.
.github/prompts/07-commit-and-pr.md	Clarifies/standardizes the canonical improvement-mode rerun marker schema.
.github/prompts/ext/cycle-rollover.md	Clarifies anchor coverage rules when `cycle_anchor=both`.
.github/workflows/news-realtime-monitor.lock.yml	Updates compiled workflow lock output for schedule and compilation changes.
.github/workflows/news-election-cycle.lock.yml	Updates compiled workflow lock output for model/compilation changes.
.github/workflows/news-year-ahead.lock.yml	Updates compiled workflow lock output for model/compilation changes.
.github/workflows/news-committee-reports.lock.yml	Updates compiled workflow lock output for compilation changes.
.github/workflows/news-evening-analysis.lock.yml	Updates compiled workflow lock output for compilation changes.
.github/workflows/news-interpellations.lock.yml	Updates compiled workflow lock output for compilation changes.
.github/workflows/news-month-ahead.lock.yml	Updates compiled workflow lock output for compilation changes.
.github/workflows/news-monthly-review.lock.yml	Updates compiled workflow lock output for compilation changes.
.github/workflows/news-motions.lock.yml	Updates compiled workflow lock output for compilation changes.
.github/workflows/news-propositions.lock.yml	Updates compiled workflow lock output for compilation changes.
.github/workflows/news-quarter-ahead.lock.yml	Updates compiled workflow lock output for compilation changes.
.github/workflows/news-translate.lock.yml	Updates compiled workflow lock output for compilation changes.
.github/workflows/news-week-ahead.lock.yml	Updates compiled workflow lock output for compilation changes.
.github/workflows/news-weekly-review.lock.yml	Updates compiled workflow lock output for compilation changes.

 | **Generated** | `YYYY-MM-DD HH:MM UTC` |
 | **Workflow** | `e.g., news-morning-propositions` |
 | **Run duration** | `minutes` |
+| **Pass-2 status** | `executed in full` *(required literal for gate compliance)* |


+## 🔁 Re-run log (improvement-mode only)
+
+> Use this exact schema on every improvement re-run. Keep the heading stable (`## Re-run log`) so the gate can validate contract compliance across all 14 news workflows.
+
+- **Re-run**: `YYYY-MM-DD HH:MM UTC` · workflow=`$GITHUB_WORKFLOW` · run_id=`$GITHUB_RUN_ID` · attempt=`$GITHUB_RUN_ATTEMPT`
+  - new dok_ids: `<count or "none">`
+  - artifacts extended: `<comma-separated list or "none — content stable">`
+  - flags closed: `<count>`
+  - vintage refresh: `<"yes" or "no, IMF WEO Apr-2026 still current">`


+  grep -qE 'Pass-2[[:space:]]+status:[[:space:]]*executed[[:space:]]+in[[:space:]]+full' "$ANALYSIS_DIR/methodology-reflection.md" \
+    || { echo "❌ methodology-reflection.md: missing canonical 'Pass-2 status: executed in full' declaration"; FAIL=1; }
+  if grep -qiE 'Pass-2[[:space:]]+status:[[:space:]]*(not[[:space:]]+executed|skipped|deferred|partial)' "$ANALYSIS_DIR/methodology-reflection.md"; then
+    echo "❌ methodology-reflection.md: Pass-2 cannot be marked not executed/skipped/deferred/partial"
+    FAIL=1
+  fi
+  if [ "${IMPROVEMENT_MODE:-false}" = "true" ]; then
+    grep -qE '^##[[:space:]]+Re-run[[:space:]]+log' "$ANALYSIS_DIR/methodology-reflection.md" \
+      || { echo "❌ methodology-reflection.md: improvement-mode requires '## Re-run log'"; FAIL=1; }
+    grep -qE 'run_id[=:][[:space:]]*'"${GITHUB_RUN_ID:-}" "$ANALYSIS_DIR/methodology-reflection.md" \
+      || { echo "❌ methodology-reflection.md: improvement-mode requires current run_id in Re-run log"; FAIL=1; }
+    grep -qE 'attempt[=:][[:space:]]*'"${GITHUB_RUN_ATTEMPT:-}" "$ANALYSIS_DIR/methodology-reflection.md" \


+    expect(template).toContain('## 🔁 Re-run log (improvement-mode only)');
+    expect(template).toContain('run_id=`$GITHUB_RUN_ID`');
+    expect(template).toContain('attempt=`$GITHUB_RUN_ATTEMPT`');


          token: ${{ secrets.GH_AW_GITHUB_TOKEN || secrets.GITHUB_TOKEN }}
          persist-credentials: false
-          fetch-depth: 0
+          fetch-depth: 1


          token: ${{ secrets.GH_AW_GITHUB_TOKEN || secrets.GITHUB_TOKEN }}
          persist-credentials: false
-          fetch-depth: 0
+          fetch-depth: 1


          token: ${{ secrets.GH_AW_GITHUB_TOKEN || secrets.GITHUB_TOKEN }}
          persist-credentials: false
-          fetch-depth: 0
+          fetch-depth: 1


          token: ${{ secrets.GH_AW_GITHUB_TOKEN || secrets.GITHUB_TOKEN }}
          persist-credentials: false
-          fetch-depth: 0
+          fetch-depth: 1


          token: ${{ secrets.GH_AW_GITHUB_TOKEN || secrets.GITHUB_TOKEN }}
          persist-credentials: false
-          fetch-depth: 0
+          fetch-depth: 1


          token: ${{ secrets.GH_AW_GITHUB_TOKEN || secrets.GITHUB_TOKEN }}
          persist-credentials: false
-          fetch-depth: 0
+          fetch-depth: 1


Initial plan

f524c95

Copilot AI assigned Copilot and pethers May 14, 2026

github-actions Bot added the size-xs Extra small change (< 10 lines) label May 14, 2026

Copilot started work on behalf of pethers May 14, 2026 23:45 View session

Copilot AI linked an issue May 14, 2026 that may be closed by this pull request

[Agentic Workflows] Fix Pass-2 budget exhaustion, scope-compression, and re-run pattern inconsistencies across news-* workflows #2492

Open

7 tasks

fix(agentic-workflows): enforce pass-2 contract and realtime coverage…

c63a247

… window Agent-Logs-Url: https://github.com/Hack23/riksdagsmonitor/sessions/e51960bd-f4d0-48d4-9b62-9767d7a0283a Co-authored-by: pethers <[email protected]>

github-actions Bot added documentation Documentation updates workflow GitHub Actions workflows ci-cd CI/CD pipeline changes testing Test coverage news News articles and content generation agentic-workflow Agentic workflow changes size-l Large change (250-1000 lines) labels May 14, 2026

Copilot AI changed the title ~~[WIP] Fix Pass-2 budget exhaustion and scope-compression across news workflows~~ Harden news workflow pass‑2/re-run contracts and rebalance schedule/model budget for reliable 60‑min runs May 15, 2026

Copilot finished work on behalf of pethers May 15, 2026 00:02

Copilot AI requested a review from pethers May 15, 2026 00:02

Merge branch 'main' into copilot/fix-pass-2-budget-exhaustion

ec20142

pethers marked this pull request as ready for review May 15, 2026 06:13

Copilot AI review requested due to automatic review settings May 15, 2026 06:13

Copilot started reviewing on behalf of pethers May 15, 2026 06:13 View session

Copilot AI reviewed May 15, 2026

View reviewed changes

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Harden news workflow pass‑2/re-run contracts and rebalance schedule/model budget for reliable 60‑min runs#2495

Harden news workflow pass‑2/re-run contracts and rebalance schedule/model budget for reliable 60‑min runs#2495
Copilot wants to merge 3 commits into
mainfrom
copilot/fix-pass-2-budget-exhaustion

Copilot AI commented May 14, 2026 •

edited

Loading

Uh oh!

github-actions Bot commented May 14, 2026

Uh oh!

github-actions Bot commented May 14, 2026

Uh oh!

github-actions Bot commented May 15, 2026

Uh oh!

github-actions Bot commented May 15, 2026

Uh oh!

Copilot AI left a comment

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

Copilot AI commented May 14, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

github-actions Bot commented May 14, 2026

🏷️ Automatic Labeling Summary

Label Categories

Uh oh!

github-actions Bot commented May 14, 2026

🔍 Lighthouse Performance Audit

Uh oh!

github-actions Bot commented May 15, 2026

🔍 Lighthouse Performance Audit

Uh oh!

github-actions Bot commented May 15, 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!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Copilot AI commented May 14, 2026 •

edited

Loading