pulumi-neo-handoff
Pass
Audited by Gen Agent Trust Hub on May 14, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local git and pulumi commands to inspect the repository environment and launch the handoff task. These operations are limited to the local system and the vendor's own verified CLI tools.
- [PROMPT_INJECTION]: The skill summarizes conversation history to provide context for the Pulumi Neo agent, creating an indirect prompt injection surface where user-provided data could influence the target agent's behavior. * Ingestion points: Conversation history and user-defined goals are aggregated as described in SKILL.md. * Boundary markers: The summary is organized with Markdown headers but lacks explicit 'ignore instructions' delimiters. * Capability inventory: The skill has access to the local shell, git, and pulumi CLI tools. * Sanitization: Content is passed as a quoted argument to the shell to prevent character expansion or command injection, though no semantic sanitization is applied to the summary text.
Audit Metadata