Unity Code Gen

Pass

Audited by Gen Agent Trust Hub on Mar 29, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by reading and processing external codebase content.
  • Ingestion points: As described in Step 2 of SKILL.md, the agent is instructed to use Glob and Grep to read the contents of .cs and .asmdef files from the local project.
  • Boundary markers: The instructions lack explicit delimiters or instructions to treat the ingested file content strictly as data, which may allow instructions embedded in code comments or strings to be interpreted as commands by the model.
  • Capability inventory: The agent has the capability to read any project file and write new source code and test files.
  • Sanitization: There is no evidence of sanitization or validation of the codebase content before it is interpolated into the prompt used for code generation.
  • [COMMAND_EXECUTION]: The skill directs the agent to execute filesystem discovery and inspection commands.
  • Evidence: The 'Etape 2 — Inspecter la codebase existante' section in SKILL.md explicitly lists Glob and Grep commands to analyze assembly definitions, namespaces, and class structures.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 29, 2026, 01:08 PM
Security Audit — agent-trust-hub — Unity Code Gen