The agent worked around a site-blocking issue (429 errors on all direct fetches) by pivoting entirely to web search, assembling a comprehensive overview of Balsam Hill from third-party comparisons, blog posts, and news articles rather than primary sources. The final answer is detailed and well-structured, but is built almost entirely from external sources (67% web search per metadata) rather than the company's own site, leaving key pricing details and product catalog specifics unverified and incomplete.

• ›All direct fetches to balsamhill.com returned 429 (rate-limit) errors, making the primary site completely inaccessible. The agent never reached the company's own product pages, pricing pages, or about section.
• ›The agent successfully reconstructed a credible answer from search results, third-party comparisons (Taste of Home, BBB, Trustpilot), and news articles (BusinessWire on bio-based polyethylene). Search snippets contained enough detail to populate most sections: what they do, who buys, sustainability claims, and Flip Tree® technology.
• ›Pricing and payment details came from indirect sources (coupon aggregators, mentions in comparisons) rather than Balsam Hill's own pricing pages. The agent had to infer a $400–600 range and note that specific product tiers were unconfirmed.
• ›The agent correctly identified unfindable information ('pricing transparency', 'exact product count', 'market share') and flagged it as gaps rather than inventing answers—a sign of good epistemic care.
• ›The site is functionally invisible to scrapers/agents; a human-readable company narrative (about page, product pages, pricing structure) exists but was completely blocked. The agent had to reverse-engineer the positioning from comparison articles instead of the company's own framing.