The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the @peakd/hive-tx-cli package via npm as its core component. This package is maintained by PeakD, a well-known service provider in the Hive ecosystem, and the repository is linked to the skill author.
[CREDENTIALS_UNSAFE]: The skill documentation describes the management of Hive Posting and Active private keys. It advises users to store these keys in a local configuration file (~/.hive-tx-cli/config.json) with restricted permissions (600) or to use environment variables, following standard security practices for CLI-based blockchain tools.
[COMMAND_EXECUTION]: The skill functionality relies on executing the hive binary with various subcommands to perform queries and broadcast operations. This is the intended primary purpose of the skill.
[PROMPT_INJECTION]: The skill processes untrusted external data by querying the Hive blockchain, which introduces a surface for indirect prompt injection. 1. Ingestion points: Data enters the context through subcommands like hive content, hive replies, and hive feed. 2. Boundary markers: The instructions do not define explicit delimiters for untrusted blockchain data. 3. Capability inventory: The skill can execute high-impact operations including hive transfer, hive publish, and hive vote. 4. Sanitization: No specific content sanitization or instruction-filtering is described in the skill markdown, though users are prompted with pre-flight checklists.

hive