Individual skill
atlas-commerce-ops-analysis
Use for Atlas Commerce Operations tasks that require producing exact JSON analytics or applying approved minimal data-quality corrections from workplace schema, dictionary, SQL data, and request payloads.
experiments/codex_gpt5_5_xhigh/reports/skills/task_group_022/reflect-3/reflect-3_attempt_02/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
- No
- Observed behavior
- Instructions only
- 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
- data Derived category
- operations Derived category
- data-databases 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
Read excerpt
# Atlas Commerce Ops Analysis
## Core Workflow
1. Read the user prompt, request payload, and answer schema before querying data.
2. Treat the answer schema as the contract. Return exactly the required JSON keys, types, rounding, array ordering, and no commentary.
3. Load schema and field definitions from the available workplace context, then identify the smallest table set needed for the request.
4. Build the cohort first. Apply business scope exactly: account production exclusions, segment/tier/region filters, warehouse/campaign/task/case windows, and inclusive or strict timestamp boundaries as stated.
5. Prefer canonical and event-derived operational truth over convenience snapshot columns when the request says cutoff, effective, final, current-at-cutoff, or corrected. Snapshot status fields can lag event history.
6. For imported source rows with `source_system`, `external_event_id`, and `ingested_at`, de-duplicate before analysis: keep the row with the latest `ingested_at`, breaking ties by the stable row id.
7. For final state at a cutoff, use effective rows with business event time at or before the cutoff, then order by event time and stable row id. Use the last state-changing row unless the request defines a different effective event.
8. If the SQL service rejects a large CTE, partitioned window, or complex join, query scoped rows in pages and aggregate locally. Always check truncation or page until complete.
9. Round only final reported values. Keep intermediate rates, currency amounts, and durations unrounded for ranking and status rules.
