v2v-install-90-day
Pass
Audited by Gen Agent Trust Hub on Jul 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is entirely instructional and template-driven. It provides prompts and structures for generating markdown documents related to organizational management based on a specific business methodology.
- [COMMAND_EXECUTION]: No shell commands, subprocess invocations, or dynamic execution patterns (such as eval/exec) were identified in the instructions or metadata.
- [DATA_EXFILTRATION]: There are no network requests, credential access patterns, or mechanisms to transmit data to external servers. The skill's scope is limited to generating text content within the agent's environment.
- [PROMPT_INJECTION]: The skill uses natural instructional language to define operational modes (Create, Update, Find). It does not contain attempts to bypass safety filters, extract system prompts, or override agent constraints.
- [INDIRECT_PROMPT_INJECTION]: While the skill can read user-provided file paths to 'Update' existing plans, it lacks the dangerous capabilities (like network access or code execution) necessary to exploit an indirect injection. The processing is limited to generating text diffs and content updates.
Audit Metadata