skill-share
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill accepts user-defined names and descriptions during the initialization process to generate skill files and Slack notifications, creating a potential surface for indirect injection if metadata contains instructions.
- Ingestion points: User-provided skill name and description during initialization (SKILL.md).
- Boundary markers: The skill does not explicitly define delimiters to separate user-provided metadata from agent instructions in the generated output.
- Capability inventory: File system write operations for project scaffolding and network requests for Slack messaging via Rube.
- Sanitization: No specific sanitization or validation logic for user-provided metadata is described to prevent instruction injection.
- [COMMAND_EXECUTION]: The skill documentation specifies a requirement for Python 3.7+ to run scripts responsible for generating the skill directory structure and assets.
- [DATA_EXFILTRATION]: The skill is designed to share metadata, descriptions, and file links to external Slack channels using Rube integrations like SLACK_SEND_MESSAGE and SLACK_POST_MESSAGE_WITH_BLOCKS.
Audit Metadata