spawn
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONPERSISTENCEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PERSISTENCE]: The skill automates the creation of agent configuration files in global directories such as
~/.claude/agents/, which are loaded across all projects and persist between sessions by design. - [COMMAND_EXECUTION]: Utilizes the
WriteandBashtools to programmatically create, manage, and verify agent prompt files on the local filesystem. - [INDIRECT_PROMPT_INJECTION]: The 'Architecture Analysis' mode ingests user-provided descriptions of technology stacks to identify relevant platforms, which represents an ingestion surface for external data.
- [EXTERNAL_DOWNLOADS]: Fetches authoritative information and authoring guidelines from official Claude documentation to ensure generated agents follow current best practices.
Audit Metadata