← Power BI DesktopCONTENT HISTORYWHAT CHANGED · RULE-BASED ANALYSIS
Update to Power BI Desktop
Snapshot Sep 30, 2026 · 23:15 UTC · version 3.0.0+codex.20260829143201
Collection source: not recorded for this historical snapshot.
First saved snapshot
No earlier snapshot is available to establish a change.
Compare saved observations
Download comparison JSONFull technical diff · 0 changed fields
Full snapshot data
{
"description": "Use when storing and reusing Power BI planning memory across cycles, including forecast versions, assumptions, actions, overrides, actuals, error causes, segment lessons, and durable learning signals for autonomous planning.",
"included_files": [
{
"relative_path": "agents/openai.yaml",
"size_in_bytes": 134
}
],
"name": "powerbi-planning-memory",
"skill_md_contents": "---\r\nname: powerbi-planning-memory\r\ndescription: Use when storing and reusing Power BI planning memory across cycles, including forecast versions, assumptions, actions, overrides, actuals, error causes, segment lessons, and durable learning signals for autonomous planning.\r\n---\r\n\r\n# Power BI Planning Memory\r\n\r\nUse this skill when the planning engine needs memory across forecast cycles.\r\n\r\n## Memory records\r\n\r\n- forecast run\r\n- scenario assumption\r\n- generated action\r\n- human override\r\n- owner response\r\n- actual outcome\r\n- model quality event\r\n- data quality issue\r\n- segment lesson\r\n\r\n## Workflow\r\n\r\n1. Preserve every planning run with timestamp, as-of date, horizon, grain, and input paths.\r\n2. Store assumptions separately from outputs.\r\n3. Link actions and overrides to later actuals.\r\n4. Generate lessons:\r\n - segment is consistently optimistic\r\n - customer converts late\r\n - product line is supply constrained\r\n - roll forecast is better than AI\r\n - human override improved the result\r\n5. Feed lessons into trust, readiness, and scenario generation.\r\n\r\n## Required outputs\r\n\r\n- `memory_id`\r\n- `event_type`\r\n- `as_of_date`\r\n- `forecast_month`\r\n- `grain`\r\n- `entity_key`\r\n- `signal`\r\n- `impact`\r\n- `lesson`\r\n- `source_artifact`\r\n\r\n## Guardrails\r\n\r\n- Keep planning memory local-first unless the user explicitly provides an approved storage path or governance workflow.\r\n- Keep memory append-only unless the user explicitly asks to clean it.\r\n- Store personal or owner-level data only when it is necessary and governance allows it.\r\n- Do not write secrets, tokens, credentials, or private contact details into planning memory.\r\n"
}SHA-256 of public snapshot: aa958673a0a7b3ea4960e91fd150b67f966f6bc6412af495d503f99793815e0a