trainingsites/campus-ai-os/competitor-analyzer

competitor-analyzer — audit the field, find the gap

Deep audit of named competitors to surface differentiation opportunities. Feeds findings into scoring-engine like the scanner does.

Kernel preamble §4 applies: references/kernel-preamble.md.

CRITICAL DIRECTIVES

• Tool-agnostic. Baseline (WebFetch/WebSearch) always works; Firecrawl multi-page crawls and Apify channel pulls light up only if present. Never assume a connector. Zero hardcoded paths.
• If a page returns an empty shell, ask for a paste or skip it — never fabricate competitor data.

Steps

1. Get the competitor list from the resolved icp context (CONTEXT_CANON (.campus-os/context-files.md, else the plugin's templates/context-files.md): founding name → canon → alias, each also matched case-insensitively before the next) or team-private atlas/config/niche.md — never hard-code shared/icp.md; accept ad-hoc names the user gives.
2. For each competitor, audit recent content: themes/topics, formats, cadence, and (where visible) engagement signals — comments, view counts, reactions. Premium: multi-page crawl + channel analytics.
3. Answer-engine cross-check (if available): ask what ChatGPT/Perplexity/Gemini surface for the niche — note where competitors own the AI answer and where it's open.
4. Identify: saturated topics (avoid), under-served angles (gap), and the differentiation wedge (what the user can own that competitors don't).
5. Hand candidate gap-topics to scoring-engine; write a competitor audit report to the reports dir.
6. Report the top 3 differentiation openings, each with one-line evidence. Offer: "Want a brief on any of these? (/atlas-daily)"