twg-space-creation
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from external repositories and existing Confluence spaces to generate content.
- Ingestion points: Reads local repository files via shell commands (
rg) and fetches existing Confluence space data using thetwgCLI. - Boundary markers: Explicitly instructs the agent in
references/proposal-flow.mdto treat embedded instructions in source material as evidence, not as authority. - Capability inventory: The skill executes
twgcommands for broad space and content management tasks. - Sanitization: The instructions strictly forbid copying secrets and sensitive data. A human-in-the-loop "Approve" stage is required before any significant mutations occur.
- [COMMAND_EXECUTION]: The skill relies on the
twgCLI tool for operation. - Context: The tool is a vendor-owned resource used for its intended purpose of infrastructure management.
- Evidence: Standard command patterns for the
twgutility are defined throughoutSKILL.mdand the reference documentation.
Audit Metadata