Skip to content

[new-pack] rabies-pep-access-global — verify complete PEP access beyond vaccine procurement - #79

Merged
cschanhniem merged 1 commit into
mainfrom
codex/rabies-pep-access-discovery
Jul 9, 2026
Merged

[new-pack] rabies-pep-access-global — verify complete PEP access beyond vaccine procurement#79
cschanhniem merged 1 commit into
mainfrom
codex/rabies-pep-access-discovery

Conversation

@cschanhniem

@cschanhniem cschanhniem commented Jul 9, 2026

Copy link
Copy Markdown
Contributor

Summary

Discovery mode. Adds public-health/rabies-pep-access-global, a scoped problem pack for verifying complete rabies post-exposure prophylaxis access rather than treating human rabies vaccine procurement as sufficient access. The pack adds five initial evidence records, a low-risk source-inventory first move, component-separation validation rules, and pack-local agent guidance.

Also repairs one source-verification failure encountered during the run by downgrading resistancemap-one-health-trust to a low-confidence One Health Trust routing-source record after the interactive ResistanceMap subdomain returned HTTP 500.

Candidate Ranking

Problem Scale (source) Tractability Neglectedness Evidence gap Why not the others
Rabies PEP access and bite-risk triage About 59,000 human rabies deaths annually; dog bites and scratches cause 99% of human rabies cases (WHO rabies fact sheet, 2024: https://www.who.int/news-room/fact-sheets/detail/rabies). Prompt PEP prevents death; Gavi opened human rabies vaccine support in 2024, but the pack can verify where vaccine supply, RIG/monoclonals, intradermal delivery, reporting, and triage are the bottleneck. WHO identifies rabies as an NTD affecting marginalized populations; Gavi support excludes RIG and dog vaccines (WHO/Gavi/UAR release, 2024: https://www.who.int/news/item/13-06-2024-gavi-to-boost-access-to-life-saving-rabies-vaccines-human-in-over-50-countries-gavi-who-and-uar). Which countries or facility tiers can convert vaccine procurement into complete, timely PEP, and where should ministries or Gavi country-support teams change procurement, training, or reporting plans? Picked: strongest combination of current allocation window and measurable component gap.
Hepatitis B birth-dose gap Hepatitis B caused an estimated 1.1 million deaths in 2024 (WHO hepatitis B fact sheet, 2026: https://www.who.int/news-room/fact-sheets/detail/hepatitis-b). Birth-dose vaccination is known and scalable. Attention and evidence pipelines are stronger through immunization systems, WHO, CDC, and national EPI reporting. Which facilities miss within-24-hour birth dose and why? Larger burden, but less missing from this repo because vaccination-coverage/equity and health-system packs already partially cover the measurement lane.
Preterm birth care quality 13.4 million babies born preterm in 2020, with nearly 1 million deaths from preterm complications (WHO/UNICEF/PMNCH news, 2023: https://www.who.int/news/item/09-05-2023-152-million-babies-born-preterm-in-the-last-decade). Kangaroo mother care, antenatal corticosteroids, and neonatal care pathways are tractable. Global newborn-health institutions already have substantial evidence pipelines. Separate incidence measurement from quality-of-care failure and facility readiness. Important, but overlaps heavily with existing neonatal/stillbirth/maternal quality packs.
Congenital syphilis screening and treatment gap 661,000 congenital syphilis cases and 355,000 adverse birth outcomes estimated in 2016 (Korenromp et al., PLOS One 2019: https://doi.org/10.1371/journal.pone.0211720). ANC screening and benzathine penicillin treatment are tractable. Resurgence is salient in some settings; global evidence is older and uneven. Which ANC systems fail at screening versus treatment versus partner management? Strong candidate, but the current global estimate is older and the pack would need more source work before beating rabies on immediacy.
Lead exposure source attribution WHO cites IHME estimates of more than 3.5 million deaths attributed to lead exposure in 2023 (WHO lead poisoning fact sheet, 2026: https://www.who.int/news-room/fact-sheets/detail/lead-poisoning-and-health). Source identification and exposure removal are tractable in some contexts. Bloomberg and WHO launched recent lead initiatives; attention is rising quickly. Which sources dominate child blood lead by locality, and which interventions are verified? Already represented by public-health/lead-exposure-urban-global; better as enrichment than discovery.

Files Changed

  • Added problem-packs/public-health/rabies-pep-access-global/ with canonical pack files plus pack-local AGENTS.md and analytic-use-note.md.
  • Updated generated routing/wiki artifacts: tasks-available.json, agent-radar.json, and docs/wiki/*.
  • Repaired source verification for public-health/antimicrobial-resistance-surveillance-global by narrowing resistancemap-one-health-trust.
  • Added a verified transport allowlist entry for the WHO NCD data portal after browser verification.

Evidence Added Or Changed

  • Added five rabies evidence records: WHO fact sheet, WHO/Gavi/UAR 2024 support announcement, WHO 2018 position paper, Sreenivasan et al. 2019 PEP logistics assessment, and WHO Rabies Modelling Consortium 2019 model.
  • Changed one AMR evidence record from an interactive ResistanceMap data-source claim to a low-confidence routing-source claim because the old URL returned HTTP 500 in verification.

Validation

  • pnpm build passed.
  • pnpm validate passed.
  • pnpm reproducibility:check passed for 549 tasks.
  • pnpm verify:sources passed: 331 evidence URLs verified, 10 existing allowlisted warnings.
  • git diff --check passed.
  • Pre-push hook reran validate and build successfully.

Known Limitations

  • This is a discovery pack, not an accepted operational claim. It should not be used for patient triage, facility ranking, or PEP substitution decisions.
  • The pack deliberately starts with source inventory because vaccine procurement, RIG/monoclonal access, bite-risk triage, patient cost, reporting cadence, and dog-vaccination context are different evidence objects.
  • No claims.json was added; future claims must include kill conditions and reviewer gates.

Reviewer Type Needed

Domain reviewer with rabies, immunization supply, One Health, emergency-care, or NTD programme experience. Red-team and field-reality review are required before any allocation or operational interpretation.

Protocol Notes

Source verification caught unrelated source rot while adding the new pack: the ResistanceMap interactive subdomain returned HTTP 500, so the evidence record was narrowed instead of allowlisted. This is the right failure mode for the protocol: stale dashboards should decay into limited routing evidence unless a reviewer can verify the underlying data path.

Daily Run Checklist

  • Read AGENTS.md, SKILL.md, SAFETY.md, all relevant schemas
  • Chose mode: discovery
  • If discovery: ranked 3-5 candidates, picked the strongest, drafted full pack, included ranking table in PR
  • Every claim has a kill condition: no new claims.json; future claim gate documented in validation.md
  • Every evidence record has dated source, stable URL, method, limitations
  • pnpm validate passes
  • pnpm reproducibility:check passes
  • pnpm verify:sources passes
  • pnpm build passes, regenerated files committed
  • PR opened with required body fields
  • Protocol Notes included

@cschanhniem
cschanhniem force-pushed the codex/rabies-pep-access-discovery branch from 64830b0 to 6dbd6e0 Compare July 9, 2026 11:43
@cschanhniem
cschanhniem merged commit 5aac538 into main Jul 9, 2026
3 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

1 participant

@cschanhniem