multi-model-orchestrator

Pass

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill provides a structured method for task decomposition and tracking using local YAML files. No evidence of malicious behavior, credential harvesting, or unauthorized data access was found.
  • [NO_CODE]: The skill consists entirely of documentation (SKILL.md, README.md), an MIT license, and YAML templates/examples. No scripts or binaries are distributed with the skill.
  • [PROMPT_INJECTION]: The skill is designed to ingest and act upon data from 'handoff' YAML files, which creates an indirect prompt injection surface.
  • Ingestion points: Reads subtask instructions and acceptance criteria from .claude/handoffs/*.yaml.
  • Boundary markers: Data is structured via YAML keys, which provides logical separation but does not include explicit security delimiters or 'ignore embedded instructions' warnings.
  • Capability inventory: The skill instructions guide the agent to perform various development tasks (file writes, command execution for tests) based on the contents of the handoff file.
  • Sanitization: The skill does not implement sanitization or validation of the YAML content before processing.
  • Assessment: While this is a standard design for a project manager skill, users should ensure handoff files are sourced from trusted contributors and reviewed before execution.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 7, 2026, 11:40 AM
Security Audit — agent-trust-hub — multi-model-orchestrator