Abm Intent Aggregator
npx machina-cli add skill akhilkannur/marketing-agent-blueprints/abm-intent-aggregator --openclawIntent Signal Aggregator
Core Instructions
You are a highly specialized AI agent focusing on Marketing Ops. Your mission is: Aggregates multiple intent data sources into a single account-level 'Heat Score'.
Implementation Workflow
Phase 1: Initialization & Seeding
- Check: Does intent_sources.txt exist?
- If Missing: Create intent_sources.txt using the sampleData provided in this blueprint.
- If Present: Load the data for processing.
Phase 2: The Loop
- Read:
intent_sources.txt. - Group: Signals by Domain.
- Score: G2=3, WebPricing=5, 6sense=4.
- Output: Save
account_heat_map.csv.
Blueprint ID: abm-intent-aggregator Source: Real AI Examples
Source
git clone https://github.com/akhilkannur/marketing-agent-blueprints/blob/main/skills/abm-intent-aggregator/SKILL.mdView on GitHub Overview
This skill consolidates multiple intent data sources into a single account-level Heat Score to help ABM teams prioritize accounts. It reads signals from intent_sources.txt, groups them by domain, applies a fixed scoring mapping, and outputs the result to account_heat_map.csv for reporting.
How This Skill Works
The agent checks for intent_sources.txt and seeds it with sampleData if missing, then loads the data. It reads the sources, groups signals by Domain, applies scores (G2=3, WebPricing=5, 6sense=4), and writes the aggregated results to account_heat_map.csv.
When to Use It
- Consolidate signals from multiple sources into a single account-level heat score for ABM prioritization
- Before sales outreach to rank accounts by heat and focus on high-priority targets
- Onboard new intent data sources by seeding or updating intent_sources.txt
- Run periodic refreshes to reflect new signals and update account_heat_map.csv
- Prepare data for leadership or marketing ops reporting on account intent health
Quick Start
- Step 1: Check if intent_sources.txt exists in the working directory
- Step 2: If missing, seed the file using the sampleData provided in the blueprint
- Step 3: Run the workflow to read sources, group by Domain, apply scores (G2=3, WebPricing=5, 6sense=4), and save account_heat_map.csv
Best Practices
- Ensure intent_sources.txt exists and is correctly formatted; seed with sampleData if missing
- Keep domain naming consistent to improve signal grouping accuracy
- Document the score mapping (G2=3, WebPricing=5, 6sense=4) for auditability
- Regularly back up account_heat_map.csv and maintain versioning
- Verify write permissions and output location for account_heat_map.csv
Example Use Cases
- Account A aggregates G2, WebPricing, and 6sense signals to yield a high heat score for prioritization
- A new intent source is added to intent_sources.txt, triggering an updated heat map on the next run
- Marketing pairs high-heat accounts with tailored ABM campaigns based on the heat map
- Seasonal campaigns use the heat map to adjust targeting during peak buying seasons
- Seed data bootstrap ensures the heat score can be computed even with an initial missing dataset
