skills/mukul975/anthropic-cybersecurity-skills/exploiting-bgp-hijacking-vulnerabilities/Gen Agent Trust Hub
exploiting-bgp-hijacking-vulnerabilities
Fail
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPRIVILEGE_ESCALATIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill downloads an installation script from
https://get.containerlab.dev, a well-known service for managing network lab environments. - [COMMAND_EXECUTION]: The installation process involves piping a remote script to
bash -c. Additionally, the skill installs packages such asbgpalerterandpybgpstreamvia standard package managers. - [PRIVILEGE_ESCALATION]: Administrative privileges (
sudo) are employed to install orchestration tools and deploy virtual routing topologies, which is a standard requirement for this type of network security testing. - [DYNAMIC_EXECUTION]: A Python script is dynamically generated and executed at runtime using a heredoc pattern to process historical BGP update records.
- [INDIRECT_PROMPT_INJECTION]: The skill processes live and historical routing data from the RIPEstat API. Although RIPEstat is a trusted organization, the skill does not implement sanitization or boundary markers when interpolating this external content into its reporting output (Ingestion points: agent.py and SKILL.md; Boundary markers: Absent; Capability inventory: Network access, local file writing, and shell command execution; Sanitization: Absent).
Recommendations
- HIGH: Downloads and executes remote code from: https://get.containerlab.dev, https://stat.ripe.net/data/rpki-validation/data.json?resource=AS65001&prefix=10.0.0.0/24 - DO NOT USE without thorough review
Audit Metadata