game-economy-inspection

Pass

Audited by Gen Agent Trust Hub on Sep 8, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the ae-cli tool to perform data operations, including metadata discovery, ad-hoc analysis, and report persistence. These operations are restricted to the domain of game telemetry and follow the tool's verified syntax.\n- [INDIRECT_PROMPT_INJECTION]: The skill manages a theoretical attack surface where malicious data could be ingested from game telemetry.\n
  • Ingestion points: Metadata and event data fetched from the game's telemetry platform via ae-cli (e.g., event names, property descriptions).\n
  • Boundary markers: The workflow includes two mandatory gates (Step 00 and Step 01) that prevent proceeding until the user has confirmed project identifiers, data mappings, and analysis thresholds via numbered text replies.\n
  • Capability inventory: Capabilities are limited to shell execution of ae-cli for data analysis and file system operations within the /tmp/game-economy-inspection/ directory.\n
  • Sanitization: The skill utilizes structured JSON definitions for all tool invocations and requires manual confirmation for any fuzzy-matched metadata, ensuring the agent does not autonomously act on untrusted strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 8, 2026, 03:22 AM
Security Audit — agent-trust-hub — game-economy-inspection