EXTERNAL TRENDS
Scout
Finds the signal before it hits the news cycle.
HackerNews, Product Hunt, GitHub Trending, Google News (for competitor brand mentions), plus any custom RSS sources you add.
Why Scout exists
Pattern detection, not news aggregation. Scout watches the edges of the AI ecosystem + your configured surfaces, and surfaces things that are genuinely new — not repackaged noise.
Routines
What Scout does every cycle.
Anthropic signal watch
Highest priority. Flags Claude + KAIROS + Anthropic signals immediately.
AI agent space monitor
New agent tools on HN / PH / GitHub trending. Flags competitive + complementary overlap.
Research signal
arXiv AI/ML papers with traction. Surfaces architectures applicable to your work.
Regulatory monitor
Weekly scan for AI regulation affecting how agents can operate.
Cross-industry pattern
Patterns in adjacent spaces (CRM, HR tech, finance) that map to AI use cases.
DATA SOURCES
- →Hacker News (Show HN, Ask HN)
- →Reddit (ML / artificial / productivity)
- →Product Hunt
- →arXiv
- →GitHub trending
- →Custom RSS feeds
MODEL USE
MODELS.classify (Haiku) for relevance · MODELS.generate (Sonnet) for synthesis findings
WHAT IT EMITS
- →Critical Anthropic signals
- →Competitive tool launches
- →Research pattern findings
- →Regulatory change findings
The rest of the team