awesome-claude-code-subagents

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill instructs installing and loading agent definition files directly from public GitHub URLs (e.g., git clone https://github.com/VoltAgent/..., curl -s https://raw.githubusercontent.com/VoltAgent/.../agent-installer.md into ~/.claude/agents/) and those markdown agent files are read as subagent prompts/roles that the system uses at runtime, so untrusted third‑party content can directly influence agent behavior.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill's installation instructions include runtime fetch-and-execute commands that download remote content which controls agents or runs code—e.g. curl -sO https://raw.githubusercontent.com/VoltAgent/awesome-claude-code-subagents/main/install-agents.sh (then chmod +x && ./install-agents.sh) and curl -s https://raw.githubusercontent.com/VoltAgent/awesome-claude-code-subagents/main/categories/09-meta-orchestration/agent-installer.md -o ~/.claude/agents/agent-installer.md which injects agent prompt content at runtime.