Menu

Individual skill

fewshot_attempt_02

0upvotes

Not available

experiments/claude_code_glm_5_2_max/reports/skills/task_group_002/fewshot/fewshot_attempt_02/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
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.

Standard SKILL.md format has not been attested at this commit. No installation command is available; inspect the source.

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.

Declared tags are partial or unavailable.

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
# SKILL: MedBridge Sales Ops — Quote & Reconciliation Decision Packages

Executable field-craft for producing account/finance-ready JSON from the MedBridge Sales Ops remote API. Two task families share one environment; the output schema is dictated **per-task** by `input/payloads/answer_template.json`. Always read that template first and mirror its exact field names, nesting, and enum values — quote tasks come in at least three template variants and reconciliation tasks in at least two. Do NOT assume a single canonical schema.

## 0. Environment & API discipline

- Source of truth = the remote HTTP JSON API. Base URL is given to the runner as `API_BASE_URL` (env file: `<remote-env-url>`). Never start a local server; never read any `env/` directory.
- Endpoints: `GET /health`, `GET /api` (metadata), `GET /api/<collection>`, `GET /api/<collection>?<key>=<value>` (case-insensitive, AND-combined, 2dp numeric tolerance, matches nested keys), `GET /api/<collection>/<id>` **detail-by-id**, `GET /api/search?q=<text>` (cross-collection substring, ≤100 hits).
- **Detail-by-id is available ONLY for:** `customers`, `products`, `rfqs`, `quotes`, `freight-quotes`, `opportunities`. For `invoices`, `payments`, `revenue-journals`, `events`, `vouchers`, `policies` use the listing endpoint with a filter, or `/api/search`.
- Canonical filter keys: `freight-quotes?quote_id=…`; `invoices?opportunity_id=…`; `payments?opportunity_id=…`; `revenue-journals?opportunity_id=…`; `events?customer_id=…` (or `?opportunity_id=…`); `vouchers?customer_id=…`; `quotes?rfq_id=…`.
- Money = USD, 2 decimals. Dates = ISO `YYYY-MM-DD`. Use stable record IDs verbatim (e.g. `Q-TR-WC-1187`, `FR-WC-AIR`, `OPP-TR-HELIOS`, `CUST-MERIDIAN`).
- The shared business/"as-of" date across tasks is **2026-06-01** (the API `generated_at` and the quote_date used throughout). Use it as `as_of_date` when a reconciliation template requires one and the prompt does not state another.
- Reconcile the prompt's named entity (customer/quote/opportunity ID) against the API; **trust the API for IDs, amounts, dates, statuses, and tier facts**. One exception for the *display name* only — see §6.
- Output: only valid JSON matching the task's template. No markdown fences, no prose. Money as numbers (cent-level precision, e.g. `42480.0`). Nulls where the template allows (`null`), not omitted.
Back to top