rt-replication-package

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read external venue policies and source maps from within the user's project (e.g., resources/official-source-map.md). This constitutes an attack surface for indirect prompt injection if those files contain malicious instructions, though in the context of academic replication, this is a standard operational requirement.
  • [DATA_EXPOSURE]: The skill proactively mitigates data exposure by mandating that confidential data never be committed and that restricted access data should be replaced with synthetic data for replication purposes.
  • [COMMAND_EXECUTION]: While the skill describes a process that involves running code end-to-end, it does not provide any hardcoded commands, shell scripts, or automated triggers that would execute without user context; it serves as a procedural guide for the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 10:29 AM
Security Audit — agent-trust-hub — rt-replication-package