subagent-delegation

Pass

Audited by Gen Agent Trust Hub on Aug 20, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill relies on the devspace CLI utility to perform subagent management, including listing, running, and continuing sessions via devspace agents commands.
  • [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface where it ingests and processes data from external subagents. If a subagent's output is compromised or contains malicious instructions, the main agent could be influenced.
  • Ingestion points: Subagent responses are retrieved using the devspace agents show command as described in SKILL.md.
  • Boundary markers: No specific delimiters or XML tags are mandated for subagent output to separate it from the main prompt context.
  • Capability inventory: The skill utilizes the devspace agents suite (ls, run, continue, show) for process management.
  • Sanitization: The instructions include a mitigation strategy requiring the agent to "inspect changed files," "run or explain relevant tests," and "verify that important claims are supported by repo evidence" before accepting subagent output.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 20, 2026, 08:28 AM
Security Audit — agent-trust-hub — subagent-delegation