From nw
Provides 5-layer testing framework for AI agents: output quality, integration validation, adversarial review, peer critique, and security checks including prompt injection resistance.
How this skill is triggered — by the user, by Claude, or both
Slash command
/nw:nw-agent-testingThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Validate agent produces correct, well-structured outputs for typical inputs.
Validate agent produces correct, well-structured outputs for typical inputs.
Test: Agent follows workflow phases | Outputs match expected format/structure | Domain-specific rules correctly applied | Token efficiency within bounds
How: Manual invocation with representative inputs. Check against acceptance criteria in agent description.
Validate correct input/output between agents in workflows.
Test: Input parsing handles upstream format | Output format matches downstream expectations | Error signals propagate correctly | Subagent mode activation works (skip greet, execute autonomously)
How: End-to-end workflow execution through full agent chain (e.g., DISCUSS -> DESIGN -> DELIVER).
Challenge validity of agent outputs rather than accepting at face value.
Test: Source verification (cited sources real and accurate?) | Bias detection (favors one approach without evidence?) | Edge case coverage | Completeness (required sections present?)
How: Peer review by -reviewer agent using structured critique dimensions.
Independent review to catch biases and blind spots in agent design.
Test: Definition follows validation checklist? | Redundant Claude default instructions? | Over/under-specified? | Could simpler agent achieve same results?
How: @nw-agent-builder validates via 11-point checklist or @agent-builder-reviewer runs structured review.
Test resilience against misuse and prompt injection.
Test: Tool restriction enforcement | maxTurns respected | Permission mode correctly scoped | Agent stays within declared scope
How: Frontmatter fields enforce at platform level. Verify configuration.
Claude Code platform provides injection resistance through: subagent isolation (own context, no sub-subagents) | Tool restriction via frontmatter tools | Permission modes via permissionMode | Hook-based validation (PreToolUse, PostToolUse)
Do NOT add prose-based injection defense. Configure platform features:
---
tools: Read, Glob, Grep # Only tools this agent needs
maxTurns: 30 # Prevents runaway execution
permissionMode: default # User approves dangerous actions
---
tools restricted to minimum necessary (least privilege)maxTurns set to prevent runaway executionpermissionMode appropriate for risk levelBash unless agent requires command executionWrite unless agent creates/modifies filesnpx claudepluginhub nwave-ai/nwave --plugin nwRuns 57 OWASP LLM Top 10 security tests against live Agentforce agents, covering prompt injection, data leakage, excessive agency, and more. Produces an A–F security grade.
Tests and benchmarks LLM agents with behavioral testing, capability assessment, reliability metrics, and production monitoring. Uses AgentBench, τ-bench, ToolEmu, and Langsmith.
Evaluates LLM agents through behavioral testing, capability assessment, reliability metrics, and production monitoring—where top agents score under 50% on real-world benchmarks.