runtime-behavior-probe

Pass

Audited by Gen Agent Trust Hub on Jul 11, 2026

Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses node:child_process.spawnSync to execute Git commands for environment metadata and uses npx to run TypeScript tools (tsc and tsx) during the probe lifecycle.
  • [REMOTE_CODE_EXECUTION]: The skill generates and executes temporary TypeScript files. It employs dynamic import() to load modules from paths constructed at runtime, enabling execution of branch-specific repository code.
  • [EXTERNAL_DOWNLOADS]: Fetches command definitions and behavioral contracts from official documentation at redis.io. The use of npx may trigger downloads of Node.js utilities from public registries.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection. 1. Ingestion points: Processes data from Redis server responses and redis.io documentation. 2. Boundary markers: No explicit delimiters are defined for untrusted external data. 3. Capability inventory: Uses high-privilege functions including spawnSync and import(). 4. Sanitization: Lacks explicit escaping or filtering logic for external content before it is included in findings reports.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 11, 2026, 03:44 AM
Security Audit — agent-trust-hub — runtime-behavior-probe