AGENT SKILL

structured-risk-evaluation

Structured tabular de-identification, release-risk evaluation, and compliance evidence.

License 许可
Apache-2.0
最近更新
2026-09-03
Stars
★ 351
Forks
27

01 前言 Front Matter

name
structured-risk-evaluation
license
Apache 2.0
metadata
{ "disco-role": "operating" }
description
Structured tabular de-identification, release-risk evaluation, and compliance evidence.
disable-model-invocation
true

02 资源文件 4

03 SKILL.md 正文

Structured Risk Evaluation

Use this sub-skill for synthetic or approved tabular release workflows that need quasi-identifier discovery, k-anonymity, l-diversity, t-closeness, differential privacy aggregates, membership risk, anonymized output validation, and compliance evidence.

Route elsewhere when the task is about

  • Free-text PHI or PII redaction → deidentification-privacy.
  • Clinical entity extraction or grounding → clinical-extraction-grounding.
  • Service deployment, transport, or adapter wiring → interoperability-serving.

Core surfaces

  • openmed.structured
  • openmed.risk
  • openmed.eval
  • openmed.compliance
  • CLI families: risk, audit, compliance, benchmark, eval, gates, calibrate

Safe workflow

  1. Discover candidate direct identifiers, quasi-identifiers, sensitive columns, and the privacy unit with scan_table or profile_structured_table.
  2. Review the policy explicitly; do not infer threshold values.
  3. Assess the raw table with assess_release, analyze_k_anonymity, assess_population_risk, or the DP and membership helpers that match the task.
  4. Anonymize with anonymize_release or anonymize_table.
  5. Re-read the materialized release and validate with validate_released_output.
  6. Package aggregate-only audit/compliance evidence and, when needed, feed it into release gates and dashboards.
  7. Keep all examples synthetic. Never depend on DUA-restricted rows, raw PHI, or deployment-time services.

Bundled references

  • references/risk-eval-workflows.md
  • references/api-reference.md
  • references/troubleshooting.md

Bundled helper script

  • scripts/structured_release_check.py

Output expectations

  • Aggregate JSON summaries only.
  • Evidence files may include Markdown and JSON, but never raw cell values.
  • Release artifacts should be materialized locally, reread, and validated before review or gating.

04 同仓库更多 Skills

VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-triton-opt

Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block/grid/warp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-solve

END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-skill

PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under ./pass_dir/, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-pattern-fusion

Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()/replacement_args()/replacement_func() files, when a pass "failed to match", or when deciding how to fuse multiple kernels/ops.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-orchestrate

Round/batch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep/revert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-feedback

Fast, deterministic feedback for PassNet work: pre-flight pattern/match verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).

★ 351