oracle

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes system-level commands and uses the vendor's 'oracle-skills' CLI to manage environment profiles and tool settings.
  • [EXTERNAL_DOWNLOADS]: It downloads and installs standard tools (Git, GitHub CLI, and ghq) from trusted sources like system package managers and reputable GitHub repositories.
  • [PROMPT_INJECTION]: The skill ingests untrusted user input via arguments for skill names and profiles, which represents a surface for indirect prompt injection. Evidence Chain: 1. Ingestion points: and arguments in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: Subprocess execution of shell commands and package managers. 4. Sanitization: No sanitization or escaping of these inputs is explicitly defined.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 01:15 PM