Individual skill
pho-algorithmic-audit
Solve Public Health Observatory (PHO) "registered algorithmic audit" analysis tasks — the ones that hand you a prompt.txt plus payloads/analysis_request.json and payloads/answer_template.json and point you at a read-only web data portal (a "<TASK_ENV_BASE_URL>" placeholder; a Public Health Observatory / Observatory data portal with /catalog, /methodology, state_health, county_health, country_indicators, socioeconomic, and revisions datasets). Use it whenever the request asks for a publication/cohort audit followed by a fixed roster of ~6 statistical "modules" (fixed-effects or GMM jackknife, nested ridge/elastic-net cross-validation, wild cluster bootstrap-t, grouped split conformal, trajectory PCA + k-means stability, source/year perturbation or Shapley, mediation, sensitivity surfaces) and a gated PASS/FAIL decision or classification, returned as one strict JSON object conforming to the answer template.
experiments/claude_code_opus_4_8_xhigh/reports/skills/task_group_023/reflect-3/reflect-3_attempt_01/SKILL.md
Source commit 5082a777379156a048881e94940d640702e5e09a
What was checked
- Matches the observed repository HEAD
- Yes
- Scanner rules are current
- Yes
- Repository scan complete
- No
- Skill scan complete
- Yes
- Observed behavior
- Network or broad shell
- Last checked
- Nov 11, 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
- data Derived category
- 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 · 6,532 bytesassets/portal.pyruns · 6,319 bytesreferences/data_model.mddoc · 5,600 bytesreferences/module_playbook.mddoc · 8,679 bytesreferences/output_contract.mddoc · 3,497 bytesreferences/release_and_cohorts.mddoc · 4,638 bytes
Read excerpt
# PHO algorithmic-audit solver
These tasks look intimidating but are highly templated. A prompt sets a business question;
`analysis_request.json` registers a cohort/publication spec plus ~6 audit **modules** and a
**decision rule**; `answer_template.json` fixes the exact response contract. Your job is to
pull evidence from the read-only portal, reconstruct each registered design exactly, and
emit one JSON object. There is **no oracle to call while solving** — you compute everything
from the portal and the request.
## What you're given (the three inputs)
- **`prompt.txt`** — the business framing and the portal base URL placeholder. Narrative
only; the real spec is in the JSON payloads.
- **`payloads/analysis_request.json`** — the authoritative spec: geography scope, years,