pp-dndbeyond
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the
dndbeyond-pp-clibinary and the@mvanhorn/printing-press-libraryfrom the author's own GitHub repository and NPM organization. These are recognized as legitimate vendor resources. - [COMMAND_EXECUTION]: The skill operates by executing shell commands via the
dndbeyond-pp-clibinary to perform rules searches and character inspections. - [DATA_EXFILTRATION]: The CLI provides a
--deliver webhook:<url>flag, allowing the agent to POST command results to an external endpoint. While a documented feature for routing output, it represents a capability for external data transfer. - [INDIRECT_PROMPT_INJECTION]: The skill processes external, user-provided data such as JSON snapshots and PDF character sheets. There is a potential risk that malicious instructions embedded within these files could influence the agent during the normalization and inspection process.
- [PERSISTENCE]: The skill implements a 'learning loop' that journals command history and synthesizes 'playbooks' locally to improve future command accuracy. This behavior is documented and data remains local to the user's environment.
Audit Metadata