跳转到主要内容
此内容尚未提供您的语言版本,正在以英文显示。

Sf Ai Agentforce Testing

技能 已验证 活跃

Agentforce agent testing with dual-track workflow and 100-point scoring. TRIGGER when: user tests Agentforce agents, runs sf agent test commands, creates test specs, validates topic routing, or analyzes agent test coverage. DO NOT TRIGGER when: Apex unit tests (use sf-testing), building agents (use sf-ai-agentforce), or Agent Script DSL (use sf-ai-agentscript).

目的

To enable thorough and automated testing of Agentforce agents, ensuring quality and reliability through detailed coverage analysis and agentic fix loops.

功能

  • Dual-track testing workflow (CLI & API)
  • Test execution and coverage analysis
  • Agentic fix loops for automated issue resolution
  • Multi-turn conversation testing via Agent Runtime API
  • Comprehensive test scenario templates

使用场景

  • Validating topic routing and action invocation
  • Analyzing agent test coverage and identifying gaps
  • Automating agent regression tests
  • Testing multi-turn conversation flows and context preservation
  • Implementing agentic fix loops for continuous improvement

非目标

  • Building or editing Agentforce agents (delegate to sf-ai-agentforce)
  • Running Apex unit tests (delegate to sf-testing)
  • Testing Agent Script DSL directly (use sf-ai-agentscript)
  • Directly managing ECA or org credentials outside the skill's context

安装

npx skills add jaganpro/sf-skills

通过 npx 运行 Vercel skills CLI(skills.sh)— 需要本地安装 Node.js,以及至少一个兼容 skills 的智能体(Claude Code、Cursor、Codex 等)。前提是仓库遵循 agentskills.io 格式。

质量评分

已验证
99 /100
1 day ago 分析

信任信号

最近提交17 days ago
星标409
许可证MIT
状态
查看源代码