blender-community-skill-bootstrap
Fail
Audited by Gen Agent Trust Hub on Aug 22, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSREMOTE_CODE_EXECUTIONCOMMAND_EXECUTION
Full Analysis
- [UNVERIFIABLE_DEPENDENCIES_AND_REMOTE_CODE_EXECUTION]: The skill directs the agent to download and install executable agent skills from a non-vendor GitHub repository (RobLe3/cc-blender-skill).
- The installation workflow utilizes a helper script
install-skill-from-github.pyto fetch and integrate external code into the agent's workspace. - Although the instructions specify a pinned commit hash (11016c9a5847897491dde935c346571bd7548e3d), the source repository belongs to an unverified individual account rather than a recognized trusted organization.
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an ingestion pipeline for external instructions that could be leveraged for indirect prompt injection attacks.
- Ingestion points: Remote skills are pulled from the
RobLe3/cc-blender-skillrepository as defined inSKILL.md. - Boundary markers: The instructions mandate that the agent informs the user of the repository source, specific revision, and license before proceeding with installation.
- Capability inventory: The installed components (e.g.,
text-to-blender,blender-modeling,blender-animationlisted inSKILL.md) grant the agent extensive control over 3D modeling environments and file export operations. - Sanitization: The workflow includes a post-installation validation step using a
quick_validate.pyscript from the$skill-creatortool to check the integrity of the downloaded folders.
Recommendations
- AI detected serious security threats
Audit Metadata