Menu

Individual skill

pho-algorithmic-audit

0upvotes

Complete registered algorithmic audits against the Public Health Observatory (PHO) data portal. Use when the user presents an analysis_request.json and answer_template.json against a PHO portal, needs to resolve publication records into analytic cohorts, run a sequence of registered statistical audit modules, and return one JSON object conforming to an answer template. Trigger on PHO, Public Health Observatory, algorithmic audit, analysis_request, answer_template, protocol_id, or any mention of a read-only health-data portal with registered audit protocols.

experiments/codex_skill_creator_comparison_deepseek_v4_pro_max/reports/skills/task_group_023/fewshot/cc/fewshot_attempt_01/SKILL.md

Read pinned source

Source commit 5082a777379156a048881e94940d640702e5e09a

What was checked

Matches the observed repository HEAD
Yes
Scanner rules are current
No
Repository scan complete
No
Skill scan complete
No
Observed behavior
Not checked yet
Last checked
Nov 6, 2026

These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.

Install this skill

Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.

An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.

Repository context

Repository stars (not skill ratings)
121
Repository last updated
Oct 29, 2026
License
Apache-2.0

Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.

Files and excerpt

File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.

Browse files
Read excerpt

# PHO Algorithmic Audit

Completes registered algorithmic audits using the Public Health Observatory portal — a read-only web portal that serves health, socioeconomic, geography, and revision data as browsable HTML tables and CSV downloads.

## Core workflow

Every audit follows the same sequence. Never skip a step or reorder modules.

1. **Read the request.** Ingest `analysis_request.json` and `answer_template.json`. The request binds entities, measures, time windows, filters, hyperparameters, random seeds, reporting precision, business gate predicates, and the controlled decision mapping. The template fixes the output contract.

2. **Access the portal.** Use the provided base URL. Preferred access is through CSV downloads (`/download?dataset=<name>&format=csv`) because they return the full dataset in one call. Fall back to HTML browsing with query parameters when filtering before download is useful, but always prefer the CSV route for analytic work. Never silently substitute rollup for direct estimates, or provisional for final releases.

3. **Resolve publications and build cohorts.** For each requested dataset, filter by the effective status, source type, value type, validity, and geography from the request. Within each entity-time-measure key select one record by the declared priority chain (highest revision, then latest release timestamp, then lowest record ID). Suppressed, invalid, withdrawn, blank, or null analytic values are unavailable and are never zero-filled. Join resolved series by stable entity-time keys, then apply completeness predicates to construct primary, balanced-panel, broad, or dual-source analytic sets. Preserve declared entity, time, feature, and group ordering everywhere.