agent-load-balancer
Pass
Audited by Gen Agent Trust Hub on Mar 1, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted task data which could contain instructions to manipulate swarm behavior or scheduling. Ingestion points: The 'tasks' array in 'scheduleWithWFQ' and 'optimizeAllocation' (SKILL.md). Boundary markers: None identified. Capability inventory: Orchestrates task migration via 'mcp.load_balance' and modifies swarm structure via 'mcp.topology_optimize'. Sanitization: No data validation or instruction filtering found in task handling logic.- [SAFE]: Algorithmic implementations for work-stealing (PriorityQueues), resource allocation (Genetic Algorithms), and circuit breaking are functionally appropriate and contain no obfuscation or suspicious file system/network activities.- [EXTERNAL_DOWNLOADS]: Operational commands reference the 'claude-flow' package which is retrieved from the npm registry using 'npx'.
Audit Metadata