Store Map
Purpose
Assemble the store side of the readiness equation: the profile plus what is physically confirmed on site, from records only. Mapping only — the readiness verdict belongs to the engine.
When to use
After pack-ingest in every run, or whenever a store's confirmed state is needed for the readiness test.
When NOT to Use
- A map of where departments sit, a floor plan or a wayfinding sign — use a store layout / planogram tool instead; this skill maps records, not space.
- The campaign pack or price-change file needs ingesting — use pack-ingest instead.
- "Are we ready" / "any gaps" — use readiness-check instead once store.json exists.
- Ranking the gaps or writing the brief — use exception-raise instead for the ranking, use brief-draft instead for the brief.
Where the data comes from
This plugin ships no sample data. Every input is the user's own Microsoft 365 content:
- Files the user names, links or attaches in the chat.
- Their OneDrive or SharePoint, found with Work IQ search (Microsoft 365 search) by campaign ID, store ID or file name when no location is given.
- Email attachments (for example a store manager's note) found the same way.
Read each source with the file reader (ReadFileContent) and cite it by file name and row or page. Write the working JSON files (store.json, shelf-prices.json) to the session working folder, never into the plugin. If a required source cannot be found, say which one is missing and ask the user where it is. Never invent it, estimate it or fall back to example values.
Inputs
Store master export (format, cluster, fixtures); receiving confirmations; task-system export; current POS shelf-price extract, from the user's Microsoft 365 content (see Where the data comes from). Keep the export identity (export_id / ASN / row) as the citation.
Steps
- Build
store.jsonfrom the exports (schema below):store_id,format,cluster,fixtures[],received_assets[],completed_tasks[], plusopen_tasks[]andcitations{}. - Build
shelf-prices.jsonfrom the POS extract: a flat{ "<SKU>": <current shelf price> }map — the exact shapereadiness_check.py --pricesreads. - Keep export identities as citations on every record; leave
prior_execution_history[]empty with a note when no history export was supplied. - Validate the merged payload if you attach it to pack.json:
python tools/validate_contract.py payload.json.
Output format
Two files, shapes fixed so the engine can read them without translation:
// store.json
{ "store_id": "ST-0781", "format": "small", "cluster": "suburban-standard",
"fixtures": ["front-window"],
"received_assets": ["SIGN-KIT-B", "POG-EC-19"],
"completed_tasks": ["hang-front-banner"],
"open_tasks": ["build-endcap", "hang-flags", "clip-strip"],
"prior_execution_history": [],
"as_of_date": "2026-08-15",
"citations": { "profile": "STOREMASTER-2026-08-15-0781", "receiving": "RECV-2026-08-15-0781",
"tasks": "TASKSYS-2026-08-15-0781" } }
// shelf-prices.json
{ "GTIN-2001": 14.99, "GTIN-2002": 7.99, "GTIN-2003": 2.49, "GTIN-2004": 4.29 }
format, cluster, fixtures, received_assets, completed_tasks are flat string lists (as in the example above). Summarise to the user as a table: field | value | source export.
Grounding requirements
Fixtures and confirmations come from records, never assumptions — "it always has an end-cap" is not a record. Every value names its export.
Guardrails
- Never fill format, cluster or fixtures from what a store "normally" has; if the store master export is missing, leave them
null, setprofile_status: "MISSING"and ask store ops for the export — readiness is blocked (references/execution-rules.md#6). - Never count an open task as done or an ASN as received without the confirmation row.
- Never include personal data (associate addresses, pay, schedules) — it is outside this contract; decline and point to HR/store ops.
- Do not compute applicability, completeness or price conflicts here; the verdict is the readiness_check engine's.
- Always show the mapped table for review before readiness-check runs.
Escalation / uncertainty
Missing store profile → confidence below the floor, readiness blocked (#6): map what the other exports support and escalate for the profile. Conflicting exports (two formats for one store) → escalate to store ops; do not pick one.