blender-animation
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill retrieves naming data from the Blender scene, such as action names, NLA track names, and strip names, which are processed by the agent. This represents an indirect prompt injection surface where maliciously named scene elements could potentially influence agent behavior.
- Ingestion points:
scripts/list_animation_actions.py,scripts/list_nla_tracks.py, andscripts/list_action_fcurves.pyread metadata strings directly from the active Blender environment. - Boundary markers: The tools do not use explicit delimiters or specific instructions to the agent to disregard instructions potentially embedded within these strings.
- Capability inventory: The skill has the ability to modify the Blender scene, delete keyframes (
scripts/delete_keyframes.py), and remove NLA tracks and strips (scripts/remove_nla_track.py,scripts/remove_nla_strip.py). - Sanitization: No filtering or validation of the retrieved naming strings is performed before they are returned to the agent context.
Audit Metadata