The Agent Skills Directory

[SAFE]: The skill operates in a read-only mode, fetching public blockchain data from established services (Hiro API and Bitflow APIs) to perform market microstructure analysis. It does not execute remote code or request excessive permissions.
[SAFE]: Data management is handled through a local cache directory in the user's home folder with a short TTL, which is standard behavior for data-heavy analysis tools to prevent rate limiting.
[SAFE]: The script uses structured JSON output and strict parsing of blockchain event logs, which minimizes the risk of indirect prompt injection from malicious contract data.

hodlmm-flow