signal-monitor

name	signal-monitor
description	Scan target accounts for new buying signals and output a prioritized alert list.

Signal Monitor

Task

Scan a list of target accounts for new buying signals across multiple sources. Output a prioritized alert list so your team acts on the hottest signals first.

This skill pairs well with /loop for scheduled, recurring monitoring (e.g., run every Monday morning).

Input Requirements

• A target account list (CSV) with at minimum: company_name, company_domain
• Signal categories to monitor (from your CLAUDE.md signal map): hiring | funding | tech_adoption | content_social | intent | competitive | timing | all
• Lookback window: 24h | 7d | 30d

Instructions

1. Load the target account list and the signal map from your CLAUDE.md
2. For each account, check for signals:
   • Hiring: New job postings matching your ICP's relevant roles (check LinkedIn Jobs, company careers page)
   • Funding: New funding announcements (check Crunchbase, press, TechCrunch)
   • Leadership changes: New executives in target departments (check LinkedIn)
   • Tech stack changes: Added or removed relevant technologies (check BuiltWith, job postings mentioning new tools)
   • Content/social: Prospect or company posted about relevant topics (check LinkedIn, company blog)
   • Intent: Website visits, G2 category research, content downloads (if intent data available)
   • Competitive: Negative competitor reviews, competitor contract renewals (if data available)
   • Web research verification: For P0 signals, use Perplexity to verify and enrich - confirm the hiring post is still active, check if the funding has a press release with details, pull recent company news for context
3. Prioritize signals:
   • P0: Act within 48 hours (high-intent signals like hiring, funding, competitor churn)
   • P1: Act within 1 week (medium-intent like tech adoption, content activity)
   • P2: Use as context in next outreach (timing signals, milestones)
4. Deduplicate: If same signal detected from multiple sources, consolidate into one entry
5. Enrich: For P0 signals, identify the best contact to reach out to and suggest an outreach angle

Output Format

Prioritized alert list (CSV or markdown):

priority, company_name, signal_type, signal_detail, source, detected_date, suggested_contact, suggested_angle, action_deadline

Summary stats:

• Total accounts scanned
• Accounts with new signals
• P0 / P1 / P2 signal breakdown

Save to: ./data/signals/signal_scan_[DATE].csv

Routing Alerts

• P0 signals: Send immediately to Slack #signals channel with: company name, signal type, suggested contact, and recommended outreach angle
• P1 signals: Include in weekly Slack digest to #signals
• Gmail notification: For teams without Slack, send a summary email to your team's shared inbox with the P0 alert list

Configure Slack webhook: set SLACK_WEBHOOK_URL in your .env file, or use the Slack MCP connector in Claude Code.

Example Usage

/signal-monitor

Input: ./data/target_accounts.csv
Categories: all
Lookback: 7d

Output:

P0 Signals (act within 48h):
- DataFlow Inc - HIRING - 3 Data Engineer roles posted 2 days ago - Source: LinkedIn Jobs
  → Contact: Alex Chen (VP Engineering) - Angle: "scaling your data team"

- Acme Corp - FUNDING - $25M Series B announced yesterday - Source: Crunchbase
  → Contact: Sarah Kim (CTO) - Angle: "post-raise infrastructure scaling"

P1 Signals (act within 1 week):
- CloudStack - TECH_ADOPTION - Added Snowflake to tech stack - Source: BuiltWith
  → Contact: Mike Johnson (Head of Data) - Angle: "Snowflake pipeline observability"

Summary: 50 accounts scanned, 8 with new signals (3 P0, 3 P1, 2 P2)

Scheduled Monitoring

To run this on a recurring schedule, use:

/loop signal-monitor --every monday --input ./data/target_accounts.csv --categories all --lookback 7d

This creates a weekly signal digest that compounds your account intelligence over time.