Menu

Individual skill

pho-algorithmic-audits

0upvotes

Solve Public Health Observatory Web-only portal tasks that provide analysis_request.json and answer_template.json and require publication/release resolution, cohort construction, registered statistical audit modules, transportability or robustness gates, and exact JSON-only answers. Use especially for exact protocol_id values PHO_STATE_TRANSPORT_AUDIT_V1, PHO_COUNTY_MEDIATION_TRANSPORT_V1, PHO_STATE_ROBUSTNESS_TRANSPORT_V1, PHO_COUNTY_PANEL_TRANSPORT_V1, or country burden revision/PCA clustering audits.

experiments/codex_skill_creator_comparison_gpt5_5_xhigh/reports/skills/task_group_023/fewshot/naive/fewshot_attempt_03/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
Yes
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 Audits

Use this skill for Public Health Observatory tasks where the prompt names a read-only Web portal, an `analysis_request.json`, and an `answer_template.json`. Produce the final response as exactly one JSON object.

## Mandatory Workflow

1. Read the prompt, `analysis_request.json`, and `answer_template.json` completely.
2. Extract the effective portal base URL from the prompt. Fetch portal evidence only from that base URL. Start with `/catalog`, `/methodology`, and `/download` if available, then use the specific data endpoints needed for the request.
3. Resolve one effective request before any data access, fit, random draw, fold split, aggregation, or decision. If the request has a known `protocol_id`, read [protocols.md](references/protocols.md) and apply the exact matching profile only.
4. Use [computation.md](references/computation.md) for release resolution, missingness, modeling, resampling, conformal, PCA/clustering, perturbation, and controlled-decision rules.
5. Build a local calculation script or notebook in the task workspace when computations are nontrivial. Keep all intermediate values unrounded and round only when constructing the final JSON.
6. Validate the final object against `answer_template.json`: required keys, field names, list lengths, enum values, identifier spelling, ordering, numeric precision, integer/Boolean/null types, and cross-field cardinalities.
7. Return JSON only. Do not include explanatory text outside the object.
Back to top