stake-game-developer

Pass

Audited by Gen Agent Trust Hub on Feb 25, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it is designed to ingest and analyze untrusted data from user-provided project directories.
  • Ingestion points: The scripts/audit-checklist.mjs script recursively walks through a target directory and reads the content of all text-based files (e.g., .md, .js, .json).
  • Boundary markers: There are no specific delimiters or instructions provided to the agent to distinguish between the script's output and the potentially malicious instructions contained within the analyzed files.
  • Capability inventory: The skill can execute local scripts via Node.js and read any file accessible to the environment. It also provides instructions to the AI based on the audit results.
  • Sanitization: The script does not perform sanitization, escaping, or filtering of the content read from the target files before presenting it as context in audit findings.
  • [COMMAND_EXECUTION]: The skill executes local Node.js scripts (validate-books-index.mjs, validate-rgs-events.mjs, and audit-checklist.mjs) to perform its primary tasks. These scripts are part of the skill's own package and operate strictly on local file data without making external network requests.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 25, 2026, 09:59 PM