Results for “black-box-testing”

Prevents AI-introduced regressions with sandbox-mode API testing, automated bug-check workflows, and patterns that catch blind spots where the same model writes and reviews code.

Use when creating test scenarios for a skill, setting up a tests/ directory for a plugin or package, running skill tests against a subagent, or verifying whether a SKILL.md changes agent behavior

Run tests and systematically fix all failing tests using smart error grouping. Use when user asks to fix failing tests, mentions test failures, runs test suite and failures occur, or requests to make tests pass.

`analysis-agent`/`task-agent`/`review-agent`: use for database, broker, cache, HTTP, framework, process, or transaction seam proof; skip local, portfolio, and release-verdict work.

当用户想测试或复盘某个 skill 的实际执行效果时使用——如“测一下这个 skill”“看看 agent 是否按 skill 执行”“生成 skill 覆盖率或溯源报告”。在隔离沙箱中发起一次 headless 盲测，逐步骤溯源并产出中文复盘报告。不要用于编写新 skill、调试业务代码、代码审查，或复盘当前会话已经发生的执行。

Integration of BDI agent architecture with soft systems methodology for complex organizational problem-solving

AIDefence security layer with prompt injection blocking, input validation, sandboxed execution, output sanitization, and STRIDE threat modeling.

Debug MarketingClaw model, provider, tool-surface, code-mode, streaming, and live/Crabbox behavior by choosing the right logs, probes, and proof path before changing code.

Audit a repo (or diff) for weak error handling — swallowed exceptions, bare catches, unhandled promise rejections, missing network timeouts/retries, and errors logged but not surfaced. Pure Claude Code, no MCP, no signup.