jarvis-core

Pass

Audited by Gen Agent Trust Hub on Apr 18, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a security-focused orchestration pipeline. Stage 0 (Input Sanitization) is designed to scan for malicious patterns such as embedded system prompts, role overrides, and tool abuse.
  • [SAFE]: Governance is integrated throughout the process. Stages 2 and 5 use policy gates to evaluate intent and plan steps against safety constraints, blocking destructive actions and ensuring least-privilege tool usage.
  • [SAFE]: The skill includes defensive measures against data exposure. The POST: Output Validator stage specifically checks for PII and hardcoded secrets (API keys, tokens) before results are presented to the user.
  • [SAFE]: The skill addresses potential indirect prompt injection surfaces through structured processing: Ingestion occurs at Stage 0/1 (user requests); Boundary markers are defined by the pipeline stages; Capabilities include orchestration of delegated tools in Stage 7; Sanitization is explicitly implemented in Stage 0 and the Post-validation stage.
  • [SAFE]: No executable code, remote dependencies, or obfuscation techniques were found. The skill operates as a structural framework for safe agent behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 18, 2026, 12:07 PM
Security Audit — agent-trust-hub — jarvis-core