ohlcv-processing

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.70). The skill's runtime scripts (scripts/process_ohlcv.py and scripts/merge_sources.py) explicitly fetch and ingest public OHLCV data from the Birdeye public API (fetch_birdeye_ohlcv calling https://public-api.birdeye.so) and accept arbitrary CSV source files, and that untrusted third-party data is then parsed and used by the pipeline to make cleaning, merging, and anomaly-driven decisions—so external content can materially influence processing behavior.