cluster-orchestration

Pass

Audited by Gen Agent Trust Hub on Jun 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent on how to manage server processes, specifically mentioning starting a background server using wsh server -L agent-$$ --ephemeral & and executing commands within sessions using terminal tools.
  • [EXTERNAL_DOWNLOADS]: The skill facilitates network communication with remote backend servers for cluster federation through the add server command. This is a core feature of the cluster orchestration model described.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it involves reading and processing data from remote terminal screens.
  • Ingestion points: Remote terminal screen content accessed via read screen operations in SKILL.md.
  • Boundary markers: Absent; the skill does not define specific delimiters to isolate remote content from agent instructions.
  • Capability inventory: The skill can execute commands in sessions (wsh_send_input), spawn background processes, and initiate network connections to new cluster nodes.
  • Sanitization: Absent; the skill does not specify filtering or validation of the screen content before it is processed by the AI.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 17, 2026, 05:50 PM
Security Audit — agent-trust-hub — cluster-orchestration