n2n-federation

Pass

Audited by Gen Agent Trust Hub on Jul 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONREMOTE_CODE_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses Python scripts (scripts/in2n-services.py) to generate and manage systemd user services, which modifies the local system's initialization and persistence layer.
  • [REMOTE_CODE_EXECUTION]: Tools such as n2n_delegate and n2n_invoke enable the agent to request execution of tasks or command-line tools on remote federated peer machines.
  • [DATA_EXFILTRATION]: The skill facilitates significant data movement through n2n_replicate (transferring RAG collections between peers) and n2n_notify_phone (pushing content to mobile devices).
  • [PROMPT_INJECTION]: The skill processes capability inventories and knowledge base responses from remote peers, which acts as a vector for indirect prompt injection.
  • Ingestion points: Remote peer capability inventories, n2n_knowledge_query responses, and n2n_task_result data found in SKILL.md.
  • Boundary markers: The skill instructs the agent to treat peer information as untrusted input, though no specific technical delimiters (like XML tags) are enforced in the prompt text.
  • Capability inventory: Remote task execution (n2n_delegate), local service management (scripts/in2n-services.py), and data replication (n2n_replicate).
  • Sanitization: Reliance on external 'model-guard' proxies and user attribution rather than input filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 30, 2026, 04:29 PM
Security Audit — agent-trust-hub — n2n-federation