x-kol-scan

Reduce Twitter API calls by using a two-pass diff scan that compares statusesCount changes per builder first, then fetches only the exact new delta tweets instead of pulling full timelines repeatedly. State tracking maintains continuity across probes.

Two-pass scanning is a data collection method where pass one probes builder profiles to detect statusesCount diffs, and pass two fetches only the newly detected posts. This stateful approach tracks last_seen_tweet_id and last_probed_at to ensure efficient delta retrieval.

Track new tweets for up to 38 builders efficiently by applying cost-aware filtering. The scan enforces diff-based fetch limits, age and type filters, and deduplicates against existing posts, ensuring you only spend API calls on actual new content.

Set up stateful Twitter monitoring by configuring the scan to maintain per-builder state variables like last_seen_tweet_id and last_statuses_count. This allows the system to calculate the exact tweet delta and execute targeted fetches rather than redundant timeline pulls.

Diff-based scanning integrates directly with scoring modules by providing filtered, deduplicated new tweets. The two-pass workflow ensures only relevant, newly detected posts are passed downstream, optimizing both API usage and scoring performance.

Using statusesCount diffs limits data collection to detecting volume changes, meaning deleted tweets or profile updates are not captured. This approach focuses strictly on identifying new posts to minimize API costs rather than tracking full timeline modifications.