Data Parrot

Explain why deals were won or lost and identify possible improvements to sales execution. Return Data Parrot's prepared closed-deal analysis grouped into win/loss reasons, sales performance factors, and buyer decision factors, with attribution, ranking, and supporting-deal evidence. Use it to identify reasons associated with the most deals or revenue. Preserve unknown reasons and distinguish attributed explanations from proven causes.

get_win_loss

Check this connection's workspace-access state, any setup or permission requirements, and available workspace choices. Use returned workspace options when the user wants to select or switch workspaces.

get_data_parrot_access_status

Discover supported fields, AI assessment meanings, filters, allowed values, and workspace-defined properties when these are unknown. Call with no arguments for the compact manifest, then request an advertised section and detail. A search requires both section and detail. This tool describes available data and how to query it; it does not return business evidence.

describe_data_catalog

Prepare a deal review or pre-call brief for one known deal ID. Return selectable HubSpot CRM fields, custom properties, associations, the seller-entered next step, future scheduled activities, and Data Parrot's AI assessment for open deals. Use its reasoning and evidence to explain forecast confidence, deal risk, and possible next actions. Use get_deal_timeline for the chronology of changes and activity. Use get_win_loss for Data Parrot win/loss reasons and evidence. Manual forecast_status is distinct from AI forecast_category. Forecast Notes contain current seller-written notes by author; get_deal loads saved history on request.

get_deal

Explain what happened on one known deal and what changed over time. Return its reverse-chronological timeline of CRM changes, Data Parrot AI assessments, and recorded customer engagements. Use it for deal history, changes in risk or confidence, and the sequence of buyer and seller activity.

get_deal_timeline

Answer sales forecast questions for a week, month, or quarter using forecast summaries, open pipeline, and won actuals. Use it for forecast reviews and team or rep comparisons. Use get_deals for individual deals behind the totals. Won actuals follow pipeline, owner, and close-date scope; open-stage filters apply only to open pipeline. Includes applicable quota targets, attainment, and gap in company currency. Use these to assess whether the team or rep is on track against the returned target. Preserve the returned owner/team, pipeline and period scope. Manual forecast_status is distinct from AI forecast_category.

get_forecast

Retrieve the next page, additional detail, or supporting evidence offered by a previous Data Parrot result. Supply its result_id and one returned continuation_id; the original query settings are preserved. Use this when the previous result offers a continuation needed to answer the question. Transcript pagination uses get_transcript instead.

expand_result

Read the recorded conversation behind a meeting or call, including buyer statements, objections, decisions, and next steps. Requires an exact transcript_id returned by get_engagements. Returns speaker turns with timecodes elapsed from recording start. If data.scope.has_more is true, continue with cursor=data.scope.next_cursor.

get_transcript

Identify customers needing attention, quiet accounts, and changes in customer health. Return exact Customer Health KPIs, paginated company records, or count groups. For ranked risk lists, state the ranking basis and describe partial rows as the highest-ranked returned accounts. Do not sum company financial values. Use get_deals for open expansion opportunities.

get_companies

Explain what grew or shrank the pipeline, which deals caused the change, and how deals moved between stages. Return pipeline history for one requested date range, including the supported waterfall, stage-transition, and deal-change views. Calendar grouping can include earlier dates; use day for exact date ranges and check scope.date_basis.range for evidence dates. For comparisons, call once per range with identical non-date arguments; align by explicit ranges and returned order, and treat unavailable history as missing, not zero.

get_pipeline_history