bt-firewall-audit
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONPRIVILEGE_ESCALATION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes various system-level diagnostic commands including
iptables,firewall-cmd,ufw, andsystemctlto retrieve firewall configurations and service statuses. - [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it ingests and parses output from local system commands (like rule lists and service statuses). However, it includes safety instructions to validate output and uses structured reporting templates to mitigate risks.
- Ingestion points: Output from
iptables,firewall-cmd, andufwcommands (SKILL.md). - Boundary markers: The skill uses structured Markdown report templates and explicit validation checklists.
- Capability inventory: File read, system command execution, and firewall rule modification via MCP tools (FirewallPortSet, FirewallIpSet).
- Sanitization: Instructions specify脱敏 (masking) of sensitive data and verification of evidence.
- [PRIVILEGE_ESCALATION]: The skill references tools (
FirewallPortSet,FirewallIpSet) that allow for the modification of system firewall rules. While these are necessary for the skill's management functions, they represent high-privilege capabilities. The skill mitigates this by requiring explicit user authorization for write operations.
Audit Metadata