signal-classifier

Classify competitor artifacts into 11 signal types with ICP scoring and urgency routing.

Updated May 23, 2026
One-click install
npx skills add https://github.com/TheGraphicKing/skott-canvas --skill signal-classifier-thegraphicking
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: signal-classifier
Source: https://github.com/TheGraphicKing/skott-canvas/tree/main/skills/signal-classifier
Command: npx skills add https://github.com/TheGraphicKing/skott-canvas --skill signal-classifier-thegraphicking

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Raw competitor artifacts scraped from the web arrive unstructured and unclassified, making it impossible to route them to the right intelligence pipeline or detect trends over time. This Skill applies a fixed taxonomy, ICP relevance scoring, and urgency flagging so every artifact becomes a consistent, comparable classification record. ## Core Features & Use Cases - Fixed Taxonomy Classification: Assigns exactly one primary signal type from 11 approved categories (product-launch, pricing-change, hiring-signal, etc.) plus up to 3 secondary types. - ICP Overlap & Impact Scoring: Scores relevance against four standard ICP personas and computes a weighted 0-10 impact score from ICP overlap, signal urgency, competitor tier, and recency. - Urgency-Based Routing: Flags artifacts as urgent, standard, or low-priority and routes them to the priority queue, batch queue, or archive accordingly. - Use Case: After a competitor-scraper run returns 40 new artifacts, run this Skill on each one to produce structured classification records, then route the two urgent pricing-change artifacts directly to the gap-analysis priority queue. ## Quick Start Classify this scraped competitor artifact using the signal-classifier skill and return the structured classification record with its routing destination.

Frequently Asked Questions about signal-classifier

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I classify competitor intelligence signals automatically?▼

Pass the structured artifact from your competitor scraper along with the competitor tier to the signal-classifier. It assigns a primary signal type from an 11-type taxonomy, scores ICP overlap, computes an impact score, and returns a routing destination.

What signal types does competitive intelligence classification support?▼

The taxonomy covers 11 types: narrative-push, product-launch, pricing-change, partnership-announcement, hiring-signal, customer-win, analyst-coverage, negative-coverage, event-activation, content-cadence-shift, and personnel-change.

How is the competitor signal impact score calculated?▼

The impact score is a weighted 0-10 sum: ICP overlap (40%), signal type urgency (30%), competitor tier (20%), and recency (10%). Each component is documented alongside the final score for auditability.

What happens when classification confidence is low?▼

When confidence falls below 0.75, the record must include an ambiguity_note explaining the uncertainty. A low-confidence classification without an explanation note fails quality review.

Can I re-classify historical artifacts after a taxonomy update?▼

Yes. The Skill supports re-classification passes on historical artifact sets when taxonomy definitions change, and accepts an optional taxonomy_version parameter to pin a specific version instead of the latest.