cli-anything-3mf
Pass
Audited by Gen Agent Trust Hub on Jun 30, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill installs the
cli-anything-3mfharness directly from the author's repository on GitHub (github.com/HKUDS/CLI-Anything.git). - [COMMAND_EXECUTION]: The agent is instructed to execute shell commands (cli-anything-3mf) to perform mesh analysis, resizing, and repair on 3MF files.
- [PROMPT_INJECTION]: The skill processes untrusted 3MF files via the
infoandinspectcommands, which return data to the agent's context. This creates a surface for indirect prompt injection if a malicious file contains instructions embedded in its metadata or mesh structure. - Ingestion points: Data extracted from 3MF files during
infoandinspectoperations. - Boundary markers: No delimiters or warnings are used to isolate file content from instructions.
- Capability inventory: The skill allows file inspection, modification, and comparison through shell commands.
- Sanitization: The skill does not explicitly mention sanitizing or validating the contents of the processed 3MF files.
Audit Metadata