Menu

Individual skill

task-group-008-fewshot-attempt-02

0upvotes

Solve private wealth advisory benchmark tasks that ask for structured JSON planning outputs using the advisory API, local request memo, and answer template. Use for Roth conversion/RMD summaries, ILIT Crummey funding cycles, GRAT versus CRAT comparisons, and combined estate liquidity action plans where source records may conflict.

experiments/codex_gpt5_5_xhigh/reports/skills/task_group_008/demonstration_skill/demonstration_skill_attempt_02/demonstration-skill-attempt-02/SKILL.md

Read pinned source

Source commit 5082a777379156a048881e94940d640702e5e09a

What was checked

Matches the observed repository HEAD
Yes
Scanner rules are current
Yes
Repository scan complete
No
Skill scan complete
No
Observed behavior
Instructions only
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.

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

# Private Wealth Advisory JSON SOP

## Workflow

1. Read only the task prompt, request memo, and `answer_template.json` supplied for the current task. Extract `client_id`, planning horizon, engagement type, required top-level keys, enum values, and date/rounding requirements.
2. Query the API, starting with `/api/health`, `/api/clients/<client_id>`, `/api/source-documents?client_id=<client_id>`, and `/api/policies/tax`. Add `/api/rmd-factors` for Roth/RMD work, `/api/retirement-accounts?client_id=<client_id>` for retirement work, `/api/life-insurance?client_id=<client_id>` for ILIT work, and `/api/trust-candidates?client_id=<client_id>` for GRAT/CRAT work.
3. Build a normalized facts sheet before calculating: controlling profile facts, account facts, insurance facts, trust candidate facts, policy constants, and the output field map.
4. Calculate with full precision, then round final USD outputs to cents. Return JSON only, with JSON numbers and booleans, not strings.
5. Set `task_id` from the task split and number, such as `train_###` or `test_###`, and set `client_id` to the stable client identifier.

Prefer the JSON API endpoints over the portal page. Use `/api/clients?search=` only when the prompt does not provide a usable client ID.

## Source Resolution
Back to top