de:wasserversorger-enrich

Journal updated. Run complete.

Enriched a 45-company batch of German water suppliers (enrichment.jsonl 1574→1619, 1614 distinct), all researched via 2 synchronous subagents, no contact made, no captchas touched. Merged cleanly (0 malformed JSON, 0 overlaps), added 19 new municipality links, ran the full map_roster.mjs→consolidate_roster.mjs→coverage.mjs pipeline — population coverage holds at 99.99% (24 residual municipalities). Noteworthy: a new TLS-expired host trap (stadtwerke-eggenfelden.de), a confirmed durable 503 trap (vgwerke-weilerbach.de), and several handelsregister/roster corrections to prior seed data, all logged in the journal for the next run.

Picked up cleanly where the 13:57 run left off (enrichment.jsonl 1574, roster.jsonl 3375, roster-links.jsonl 18739 — matched the prior entry's reported finish exactly).

Batch build: build_enrich_batch_excl.mjs 45 enrich-20260903-1530 — 1826 pending (excluding 6 standing duplicate-mint pairs), 0 pending_domain. Split 23/22, dispatched to 2 synchronous general-purpose Agent subagents (run_in_background: false, launched together, both told not to nest further sub-agents — neither did; both also told to skip Part 3 entirely when a company_id already has roster-links coverage, per the 13:57 run's lesson). Each wrote its own enrichment.part-shard{A,B}-enrich-20260903-1530.jsonl / roster-links.part-shard{A,B}-… shards.

Verified before merge: 0 malformed JSON across all 4 shard files, both shards' company_id sets matched their dispatched batch files exactly, 0 overlap between shard A and B, 0 overlap with the existing 1574-line enrichment.jsonl.

• Shard A (23): 22/23 email (1 miss: wv-de-3932 Eggenfelden, Impressum email fully JS-obfuscated with no decodable static value — recorded not_found), 23/23 phone, 20/23 emergency_phone_number (1 earned not_published, 2 not_found), 18/23 analysis_url (16/18 multi_zone), 135 fetches / 10 searches. Roster-links: 20/23 already covered and skipped per instructions; 2 companies got 17 genuinely new rows (ZVWV Pirna/Sebnitz wv-de-8513: 15 missing members added; Stadtwerke Eggenfelden wv-de-3932: 2 new).
• Shard B (22): 22/22 email, 22/22 phone, 18/22 emergency_phone_number (rest not_published/ not_found), 19/22 analysis_url (6/19 multi_zone — lower than usual, batch skewed toward small single-town Eigenbetriebe), 103 fetches / 22 searches. Roster-links: 20/22 already covered; 2 new rows (Achterwehr via WBV Rumohr; Ludwigsstadt, Frankenwaldgruppe's 7th member).
• Combined: 45/45 researched, no contact made, no captchas touched, 19 new municipality links (both shards mostly hit companies whose links were already sourced from a prior discovery pass — consistent with 13:57's shard A finding, now the normal case at this stage of the campaign).

Merge: enrichment.jsonl 1574 → 1619 lines / 1614 distinct company_id (+45 clean; the 5 line/distinct gap is the pre-existing standing duplicate-mint pairs). Roster-links merge used the dedup-before-append check (key = company_id|raw_name): roster-links.jsonl 18739 → 18758 (+19, 0 duplicates skipped this time — both shards' new rows were genuinely new). map_roster.mjs ran clean (usual crop of no_ags_for_this_name unmatched Ortsteil/Verbandsgemeinde-composite names, nothing new). consolidate_roster.mjs: 177 shards / 3617 rows, still 3375 distinct companies; roster-links.csv now 15860 distinct (+19 net). coverage.mjs: population 99.99% → 99.99% (unchanged, expected for an enrich-only pass), 24 residual municipalities / 11,023 people, unchanged.

New findings, recorded in enrichment notes:

• wv-de-2707 (Wasserverband Peine) — seed's own notes made it look like a copy-paste mismatch (a "Samtgemeinde Dransfeld" member ~80km from Peine); re-verified on the operator's own site — not an error, WVP genuinely spans ~1600km² non-contiguous, Dransfeld is one of 10 named Versorgungsgebiete (26 zones, richest analysis page in the batch). Worth remembering so a future pass doesn't "fix" this into an error.
• New TLS host trap: stadtwerke-eggenfelden.de has an expired certificate, needs curl -k (same class as bodensee-wasserversorgung.de/kelsterbach.de/stadtwerke-kronach.de/-halver).
• vgwerke-weilerbach.de — now a durable trap, 503 on every path for the 2nd consecutive observation (first seen 2026-08-20); also has its contact email on a different domain (vg-weilerbach.de) than its own site, a domain-collision-flavored quirk worth a human's eye.
• Bulk-supplier gap: wv-de-3608 (Burbach) and wv-de-3928 (Hilchenbach) both source 100% from Wasserverband Siegen-Wittgenstein, whose own site (wvs.nrw) is a JS SPA with no accessible town-specific analysis page — WVS itself is worth a follow-up enrichment pass.
• Third-party analysis-portal pattern: gw-manager.com, wasserportal.info, wasserqualitaet-online.de each appeared as the analysis host for a different company in shard A — recurring outsourced quality-reporting SaaS, possibly worth its own tier/category if it keeps recurring.
• Krumbach: Attenhausen/Edenhausen/Krumbad, inside Krumbach's own municipal area, are actually served by a separate Zweckverband Stauden-Wasserversorgung — possible roster double-claim worth checking.
• Neutraubling: no water section anywhere on the city site; self-supply status confirmed only via a 2025 lead-contamination news article, REWAG considered and ruled out as operator — flagged for follow-up.
• Handelsregister corrections to seed: wv-de-4543 Löbau (HRB 4066 Dresden), wv-de-3933 Bad Camberg (Registernummer 2572 AG Limburg — unusual for a nominal Eigenbetrieb, worth a human check on register type), wv-de-926 Sprendlingen-Gensingen AöR (HRA 41679) all had seed's handelsregister_evidence wrong or unchecked; corrected this pass.
• No non-suppliers found in this batch — all 45 are genuine, currently active drinking-water suppliers.
• pdftotext/python3 still not present in this environment; both shards worked around it (raw HTML/zlib extraction, or corroborating via a live HTML analysis page when PDF extraction failed).

Still stale: export_csv.mjs still reads the dead legacy companies.jsonl/links.jsonl pair, not roster.jsonl/roster-links.jsonl — skipped again in favor of the live pipeline, as every recent run has done. Still worth a maintainer fixing (10+ consecutive runs now).

For next run: continue with build_enrich_batch_excl.mjs 45. 0 pending_domain. Open human follow-ups (compounding across runs, none touched this pass unless noted): (1) wv-de-20661 Gauting → Würmtal-Zweckverband re-key; (2) wv-de-20662/wv-de-3587 Gilching duplicate; (3) wv-de-3751 Heilbad Heiligenstadt likely non-supplier / EW Wasser GmbH may be the real operator; (4) export_csv.mjs pointed at dead source files; (5) wv-de-16016 Ainring and wv-de-3724 Prenzlau self_only roster flags contradicted by their own sites; (6) wv-de-3763 Oberasbach RangauWerke freshness gap; (7) stadtwerke.freilassing.de HTTP 500 — retry, don't trap yet; (8) wv-de-3800 Sinzig → Stadtwerke Rhein-Ahr AöR merger; (9) wv-de-3815 Kronach HRB conflict, blocked on stadtwerke-kronach.de TLS trap; (10) wv-de-3580 Hille likely duplicate/mis-key with WBV Hartum; (11) wv-de-1502 WZV Woringen Ungerhausen membership contradiction; (12) wv-de-3825 Halver/Kierspe "Wasserwerk Jubach" naming coincidence; (13) wv-de-4404/wv-de-4403 Römerberg-Dudenhofen/Dudenhofen domain-collision, double-corroborated; (14) wv-de-3884 Grefrath 403 host trap; (15) wv-de-3876 Raunheim CMS migration, verify once the new site settles; (16) Wasserverband Siegen-Wittgenstein (WVS) itself needs enrichment — bulk supplier for Burbach/Hilchenbach, own site has no accessible analysis page (new this run); (17) Krumbach vs Zweckverband Stauden-Wasserversorgung possible double-claim on Attenhausen/Edenhausen/Krumbad (new this run); (18) Neutraubling self-supply status unconfirmed beyond a news article (new this run); (19) vgwerke-weilerbach.de durable 503 trap + cross-domain contact email (new this run).