dnd-create-monster
Pass
Audited by Gen Agent Trust Hub on Sep 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied monster descriptions and external JSON data, creating an attack surface where malicious instructions could be embedded in data.
- Ingestion points: Workflow Step 4 in SKILL.md reads user-supplied concepts; benchmark.py reads local and remote JSON compendium files.
- Boundary markers: No specific delimiters or safety instructions are defined for the interpolation of user-supplied data.
- Capability inventory: The skill executes a local Python script (scripts/benchmark.py) and generates Markdown/JSON stat blocks.
- Sanitization: No explicit sanitization or filtering is performed on external or user-supplied content.
- [COMMAND_EXECUTION]: The workflow in SKILL.md directs the agent to execute a bundled Python script (scripts/benchmark.py) using python3 to perform benchmarking and validation tasks.
- [EXTERNAL_DOWNLOADS]: The benchmark.py script is designed to fetch reference bestiary data from a remote URL (https://raw.githubusercontent.com/SirDarcanos/openfray/main/public/compendium/srd-creatures.json) if local data is not present. This source belongs to the skill author.
Audit Metadata