artcraft-cli-plan
Pass
Audited by Gen Agent Trust Hub on Oct 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted JSON data (workflows, command plans, and briefs) that influences the behavior of the agent and orchestration scripts.
- Ingestion points: Workflow JSON files (e.g.,
examples/brand-campaign.json) and user-provided requirements documents (references/brief.md). - Boundary markers: Absent; the skill does not specify delimiters or instructions to ignore embedded commands within processed data.
- Capability inventory: Execution of native design commands and file system operations via the orchestration scripts in the harness skill.
- Sanitization: The documentation does not specify sanitization or validation of the content within processed files beyond schema conformity.
- [EXTERNAL_DOWNLOADS]: Fetches Whisper speech recognition models from Hugging Face, as detailed in the mixed ASR guide (
references/mixed-asr.md). - [COMMAND_EXECUTION]: Runs Python-based orchestration and setup scripts (
bootstrap.py,workflow.py,cli.py) to manage project lifecycles and interact with domain-specific design tools.
Audit Metadata