setup-project
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses
npx github:timschoch/skillyto download and execute setup modules directly from the author's GitHub repository. This is a vendor-owned resource used for just-in-time tool installation. - [COMMAND_EXECUTION]: The instructions direct the agent to execute shell commands such as
npxfor package management andghfor triggering and monitoring GitHub Actions workflows. - [INDIRECT_PROMPT_INJECTION]: The workflow involves reading and following instructions from sub-skills installed at runtime, which creates a surface where instructions from those files influence the agent's actions.
- Ingestion points: Dynamically installed
SKILL.mdfiles from modules likesetup-pre-commitorsetup-repo. - Boundary markers: The agent is explicitly told to "READ the installed SKILL.md and follow it directly."
- Capability inventory: The skill environment has permissions for command execution and repository modification.
- Sanitization: Content from the dynamically loaded skills is processed as instructional markdown without specialized validation.
Audit Metadata