The Agent Skills Directory

[SAFE]: No malicious patterns, obfuscation, or unauthorized data access were detected. The skill operates locally through the Heptabase CLI and follows established practices for tool-based knowledge management.
[COMMAND_EXECUTION]: Shell command execution is properly restricted to the 'heptabase' and 'jq' binaries. These tools are used for their intended purposes: managing local knowledge base content and parsing JSON output.
[PROMPT_INJECTION]: No evidence of prompt injection or attempts to bypass safety filters was found. The skill includes clear instructions for the agent to use only the official CLI for data access, which maintains a secure boundary with the local application.

heptabase-cli