[gniewkob]

Closes #152 after exact-HEAD source, governance, CodeQL and full-regression acceptance, then postmerge main.

Delivered — actual official, free, public Polish 10Y-ish sovereign yields

Eurostat original dataset irt_lt_mcby_m (monthly PL, MCBY convergence criterion; percent per annum, representative central-government bonds with approximate ten-year maturity). Official detailed ESMS methodology sections 3.4, 4, 14.1, 17, 18; Eurostat reuse notice sections 2–3, rights exceptions apply.

Actual original public response: source CI #43325200200 59/59 actual observations from Nov-2021 through Sep-2026 (no missing months), original response sha256 acc302b641ca981b5c8e22259a613d092e5c12d0cb01e39e7e093c07c59c3abf, current provider update 2026-11-12T11:00:00+0100; mean yield level 5.6115% annualized, 3.12% min, 7.82% max and 154bps maximum month-to-month yield level difference — these are NOT investment returns.

Files

• scripts/eurostat_pl_bond_yield_source_probe_v04.py: bounded fixed official HTTPS JSON-stat source metadata only, no redirects, timestamp and category sanitizer, SHA.
• scripts/eurostat_pl_government_bond_yield_context_v04.py: schema-typed 59-month ORIGINAL source-verified rate parse, annual-percent rate validation, min/max/mean and source fingerprint; no row data printed.
• tests/test_eurostat_pl_*_v04.py: 19 negative/shape contracts on each Ubuntu/macOS with sockets blocked.
• docs/ISL_EUROSTAT_PL_SOVEREIGN_YIELD_REFERENCE_V04_20261112_PL.md: exact primary-source sections, rights caveats, real observations and PIT restrictions.
• Workflow source one-off / full repo regression / G1 and CodeQL paths updated; no schedule, DB access, orders, trading or new rights.

Acceptance

• Official publisher real 59/59 complete, versioned SHA source receipt; only metadata/aggregates emitted
• 19 local negative tests Ubuntu and macOS, no network
• Full exact-HEAD repo regression 1,348 OK, 1 skipped: Actions #43325692721
• GitGuardian + CodeQL static check
• Semantic/manual code inspection with exact evidence PR comment, not formal independent human approval
• G1 governance and combined-stack macOS jobs (queued; exact SHA), then merge only after all green
• Verify exact-main postmerge full regression SUCCESS and issue Public-first v0.4: original Eurostat Poland sovereign 10Y yield context (not total return) #152 CLOSED

historical_pit_certified=false / bond_holding_period_return_observed=false / qualified_model_series=[] / INSUFFICIENT_DATA_HOLD_CASH, no real financial operations, no automatic source-rights admission. #116 remains OPEN for actual Polish total-return rights/PIT.

[chatgpt-codex-connector]

Codex Review Summary: ❌ Failed

This comment shows the latest Codex review activity on this pull request.

Review	Status	Commit	Review trigger
📝 Code Review	❌ Failed 2026-11-12T23:06:12.867332Z We couldn’t complete this review. Comment @codex review to try again.	f2aa795	Draft marked ready

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review" or "@codex security review".

Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings.

[chatgpt-codex-connector]

You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard.
To continue using code reviews, you can upgrade your account or add credits to your account and enable them for code reviews in your settings.

[gniewkob]

Public-first v0.4 — source-backed review & acceptance evidence (2026-11-12)

Exact PR #153 HEAD: f2aa795693872a6d37acce7d1a36d3bb8df4dfab. Original public Eurostat irt_lt_mcby_m, freq=M/int_rt=MCBY/geo=PL — Polish secondary-market sovereign yield in % p.a., near-10Y maturity. NOT holding-period or total-return, no TBSP/PL_BONDS_TR_PLN substitution.

Actually verified publisher response: one-shot official source CI #43325692714 SUCCESS; 59/59 non-missing 2021-11…2026-09 original JSON-stat data points; raw SHA256 acc302b641ca981b5c8e22259a613d092e5c12d0cb01e39e7e093c07c59c3abf. Source metadata updated=2026-11-12T11:00:00+0100. Receipt aggregate source SHA 4157225ea7cc1f2a5bf9816f9141ab036f098a29f0d5e5f3cfecdda42388b9f6. Yields min 3.12%, max 7.82%, mean 5.6115% p.a. Max absolute month-to-month yield LEVEL change 154.00 basis points; these metrics are NOT monthly bond returns, investable profits or actual losses. source_statistical_flags_count={}.

Review performed: exact-head semantic/manual source-code inspection of source allowlist, no redirects/max response bytes, fixed (freq/int_rt/geo/time) identity and dimensions, 59 contiguous months and value units/range, nonfinite/zero/missing handling, future source window, no historical PIT assumptions, provider-update log injection, source bytes SHA and independent receipt hash, untrusted category codes, typed yield vs realized total return. No P0/P1 correctness flaw found in this source-scoped educational module. Manual assistant review is not independent human/security-owner approval. CodeQL is an independent automated static scan but does not substitute for that approval.

Actual test evidence: offline no-network 7+12=19 tests on Ubuntu and macOS both PASS; exact-head full regression #43325692721 1348 tests OK, 1 skipped. CodeQL #43325692717 and GitGuardian PASS. G1+stack Ubuntu PASS; G1+stack macOS still queued at timestamp of this comment and must pass before merge. See Polish source-referenced methods and actual evidence.

Boundary: Eurostat ESMS §3.4/4/14.1/17/18 describes percent p.a. yield, transmission and potential revisions. General Eurostat reuse policy allows statistical-data reuse with attribution subject to exceptions; not independent authorization for other vendors or financially executable activity. historical_pit_certified=false, qualified_model_series=[], legacy_m1_m2_m3=INSUFFICIENT_DATA_HOLD_CASH, zero trading/wallet/DB/schedules. Source-rights+TR+PIT #116 remains OPEN.

[gniewkob]

FINAL — v0.4 Polish official Eurostat bond yield research ACCEPTED (not bond total return)

DONE. PR #153 merged (squash) from exact PR HEAD f2aa795693872a6d37acce7d1a36d3bb8df4dfab to main dbb601746df2301d269f591573e4732a9d9765b2. Issue #152 CLOSED. Exact-main push CI #43328900483 SUCCESS, 1348 tests OK, 1 skipped; main SHA independently checked dbb601746df2301d269f591573e4732a9d9765b2.

Cancelled job triage: stack workflow #43325692736 attempt 1 was CANCELLED for queued stack-macos, while Ubuntu passed. G1 audit-contract (macos-15) actually finished SUCCESS on original attempt #43325692708. Re-ran only cancelled job via GitHub Actions connector (job 131557455678). Attempt 2 PASS both macOS and Ubuntu, no code or source changes. All 11/11 exact-HEAD checks SUCCESS including CodeQL, GitGuardian, both OS source-contract tests, full regression, official original public source fetch, G1 and stack.

Original one-shot official Eurostat irt_lt_mcby_m, PL/M/MCBY, real 59/59 months Nov2021–Sep2026, source SHA acc302b641ca981b5c8e22259a613d092e5c12d0cb01e39e7e093c07c59c3abf; yield % p.a. min 3.12, max 7.82, average 5.6115, largest adjacent-month yield level change 154 basis points, not bond holding-period investor returns. Source public-first, receipt observed 2026-11-12 not original historic first release.

No promotion: PL_BONDS_TR_PLN and M1–M3 remain INSUFFICIENT_DATA_HOLD_CASH; #116 separately remains OPEN for total-return source identity/rights and history publication PIT; no paid APIs, wallet, brokerage, DB writes, schedules, transfers or financial operations.