pipeline-review

What it does

Pulls deal and meeting data from whatever system the user tracks their pipeline in, analyzes the pipeline over a chosen time period, and produces a report that answers: Are we booking enough? Are they qualified? Where are deals getting stuck? What's working and what isn't?

Two output modes:

• Executive summary: 1-page snapshot with numbers, trends, and red flags.
• Detailed diagnostic: Full data tables, stage-by-stage breakdown, source analysis, stuck deal list, and specific recommendations.

Both are always produced. The executive summary sits at the top of the report.

How it works

The skill defines a multi-step workflow:

• On first run, perform a one-time configuration to collect preferences including data source (crm_tool), access method (access_method), and deal object name (deal_object), as well as stage definitions (pipeline_stages, qualified_stage, won_stage, lost_stage) and cycle timing (expected_cycle_days).
• Pull pipeline data via Step 1: Pull Pipeline Data, retrieving current period and an optional comparison period from the configured CRM or source, using the input contract to specify period and access details.
• Normalize all data into a standard structure with fields such as id, name, company, stage, owner, source, created_date, close_date, last_activity_date, meeting_date, amount, qualification_status, loss_reason, days_in_current_stage, total_age_days.
• Produce an output pipeline_data object containing current_period and optional comparison_period datasets, including deal counts.
• Run Step 2: Analyze Pipeline across seven dimensions (Volume, Qualification, Source Attribution, Stage Distribution & Velocity, Stuck/At-Risk, Win/Loss), applying period-over-period comparisons when a comparison period exists.
• For each analysis, compute the specific metrics described in the skill, producing structured outputs such as volume metrics, qualification breakdown, source attribution, stage analysis, stuck deals, and win/loss results.
• Present a human checkpoint summary indicating source, date ranges, counts, and field availability, prompting user confirmation.

When to use it

Load this composite when:

• The user says phrases like "review my pipeline", "pipeline report", "how's our pipeline looking", or "deal review".
• The user indicates 1:1 prep, board meeting prep, or weekly sales review.
• An upstream workflow triggers an end-of-period review.
• The user asks about meeting quality, conversion rates, or pipeline health.

What it can touch

The skill intends to pull data from CRMs and spreadsheets and normalize into a standard structure, including fields like created_date, close_date, amount, stage, source, and meeting_date. It uses the following inputs and mappings:

• crm_tool, access_method, field_mapping (from Step 0: Configuration)
• pipeline_stages, qualified_stage, won_stage, lost_stage, expected_cycle_days
• benchmark targets (optional): target_meetings, target_qual_rate, target_win_rate, target_pipeline_value
• standardized deals array in pipeline_data.current_period.deals and pipeline_data.comparison_period.deals

Caveats

The skill aims to be tool-agnostic, working with various CRMs (Salesforce, HubSpot, Pipedrive, Close, Supabase, CSV, Google Sheets, Notion, etc.). Data quality limits exist: missing fields adaptively reduce analysis detail; some defaults (e.g., stage time thresholds) are provided for estimation when data is sparse.