Individual skill
pho-registered-audits
Solve Public Health Observatory registered audit tasks that use a TASK_ENV_BASE_URL portal plus analysis_request.json and answer_template.json. Use this skill for state, county, or country Observatory algorithmic audits involving release/cohort resolution, fixed effects, GMM, ridge or elastic-net validation, wild cluster bootstrap, conformal calibration, PCA clustering, source perturbation, sensitivity surfaces, or controlled JSON decisions.
experiments/codex_skill_creator_comparison_gpt5_5_xhigh/reports/skills/task_group_023/fewshot/opencode/fewshot_attempt_02/SKILL.md
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.
- agent Repository topic
- ai Repository topic
- benchmark Repository topic
- llm Repository topic
- self-evolving Repository topic
- self-learning Repository topic
- skills Repository topic
- security Derived category
Files and excerpt
File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.
Browse files
SKILL.mdcore · 3,924 bytesSKILL.mdcore · 3,924 bytesSKILL.mdcore · 3,924 bytesSKILL.mdcore · 3,924 bytesreferences/methods.mddoc · 15,406 bytesscripts/pho_audit_utils.pyruns · 24,395 bytes
Read excerpt
# Public Health Observatory Registered Audits
## Operating Rule
Use only the future task's prompt, `analysis_request.json`, `answer_template.json`, and the authorized Public Health Observatory portal. Do not reuse values from examples or prior runs. The template controls the final JSON shape, key names, enum values, ordering, booleans, and numeric precision.
Read [references/methods.md](references/methods.md) before computing modules. Import [scripts/pho_audit_utils.py](scripts/pho_audit_utils.py) when deterministic portal download, release selection, matrix algebra, bootstrap PRNG, PCA, k-means, conformal ranks, or rounding helpers would reduce implementation risk.
## Workflow
1. Read the prompt and both payload files completely. Replace `<TASK_ENV_BASE_URL>` with the actual base URL from the task environment.
2. Freeze one effective request before data access. Apply direct request keys and any `*_overrides` by exact-path deep merge: objects merge recursively, arrays replace whole arrays, scalars replace exact paths, and unknown targets are errors.
3. Download evidence from the portal through `/download?dataset=<name>&format=csv` plus query filters. Use `/catalog` and `/methodology` only to confirm schema, filters, current policy, and identifier semantics.
4. Resolve releases independently for each requested geography, year, measure, value type, source, and status. Select by the effective revision priority; treat suppressed, invalid, withdrawn, blank, and null analytic values as unavailable, never as zero.