skills/ruvnet/ruflo/cost-export/Gen Agent Trust Hub

cost-export

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEDATA_EXFILTRATIONPRIVILEGE_ESCALATIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill is designed to transmit internal cost-tracking records from the AgentDB to external URLs via a user-specified webhook argument. This provides a mechanism for exporting data to non-whitelisted external domains.
  • [PRIVILEGE_ESCALATION]: The skill documentation suggests writing Prometheus metrics to /var/lib/node_exporter/textfile_collector/. On standard Linux systems, the /var/lib/ directory is restricted and typically requires root or specific service user privileges to write files.
  • [COMMAND_EXECUTION]: The skill executes a local Node.js script located at plugins/ruflo-cost-tracker/scripts/export.mjs using the Bash tool to perform the export operations.
  • [INDIRECT_PROMPT_INJECTION]:
  • Ingestion points: The skill processes session-* and budget-config-* records retrieved from the cost-tracking namespace in AgentDB.
  • Boundary markers: None are specified to delimit the retrieved database content during processing or transmission.
  • Capability inventory: The skill has the ability to perform network POST requests (via webhooks) and local filesystem writes based on instructions contained within its arguments.
  • Sanitization: There is no evidence of sanitization or schema validation performed on the database records before they are exported to external systems.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 11:37 AM
Security Audit — agent-trust-hub — cost-export