workspace-topology
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes project configuration from AGENTS.md and .gitmodules to manage the workspace topology. This ingestion of external configuration creates a surface for indirect instructions, although the skill explicitly implements mitigation logic to prevent path traversal or escape beyond authorized working directories.
- Ingestion points: AGENTS.md (workspace frontmatter), .gitmodules.
- Boundary markers: The protocol mandates validation rules to check if resolved paths escape the authorized working area.
- Capability inventory: Employs Bash, Read, Write, and Edit tools to perform repository management and structural modifications.
- Sanitization: Includes explicit safety rules in the protocol (protocol.md §3) to reject paths that escape authorized scopes.
- [COMMAND_EXECUTION]: The skill uses the Bash tool to execute standard git commands for repository maintenance, migration between layouts, and commit management. Safety protocols are defined to block high-risk states, such as preventing commits on a detached HEAD, and the skill requires user confirmation for destructive actions or network operations like pushing code.
Audit Metadata