skills/daxaur/openpaw/c-network/Gen Agent Trust Hub

c-network

Pass

Audited by Gen Agent Trust Hub on Mar 9, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides terminal commands for DNS lookups and HTTP requests using the doggo and httpie utilities. These commands allow the agent to interact with external network services via the command line.
  • [PROMPT_INJECTION]: The skill facilitates an indirect prompt injection surface by processing data from remote network sources.
  • Ingestion points: Data enters the agent's context through output from doggo (DNS records) and httpie (HTTP response headers and bodies).
  • Boundary markers: There are no instructions to wrap these outputs in delimiters or use specific system instructions to prevent the agent from interpreting network data as instructions.
  • Capability inventory: The skill includes commands for querying DNS servers and performing a wide range of HTTP operations (GET, POST, file downloads, authentication).
  • Sanitization: There is no evidence of sanitization, filtering, or validation applied to the external data before it is ingested by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 9, 2026, 08:44 AM
Security Audit — agent-trust-hub — c-network