gsd-next
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted project data and user-provided arguments. Ingestion points: The skill interpolates '$ARGUMENTS' into the prompt context and reads project/workflow state using 'gsd-tools smart-entry --json' as described in 'SKILL.md'. Boundary markers: No explicit delimiters or instructions are provided to the agent to treat external project data or arguments as untrusted content. Capability inventory: The skill uses 'Bash', 'Read', and 'SlashCommand', providing a significant capability surface if the agent is manipulated via malicious data. Sanitization: No sanitization or validation logic is defined for the external inputs.
- [COMMAND_EXECUTION]: The skill executes a local command 'gsd-tools smart-entry --json' to determine the project state. This execution is part of the intended primary functionality of the skill as a state-aware router.
Audit Metadata