cxas-agent-foundry

分类: 工具与效率 | 上传者: GoogleCloudPlatform | 下载: 0 | 版本: v1.0（最新）

端到端 GECX/CXAS/CES 对话式代理生命周期——根据需求构建代理（PRD 到代理）、创建和运行评估（黄金、模拟、工具测试、回调测试）、调试故障并迭代到生产质量。每当用户提及 Google Customer Engagement Suite 平台上的 GECX、CXAS、CES、SCRAPI、对话代理、语音代理、音频代理、代理评估、推/拉/linting 代理或代理指令/回调/工具时，请使用此技能。

更新日志: Source: GitHub https://github.com/GoogleCloudPlatform/cxas-scrapi

目录结构

当前层级: tree/main/.agents/skills/cxas-agent-foundry/

📁 assets/
- 📁 project-template/
  - 📁 cxas_app/
    
    📁 Sample_Support_Agent/
    
    📁 agents/
    
    📁 root_agent/
    
    📁 after_model_callbacks/
    
    📁 after_model_callbacks_01/
    
    📄 python_code.py 5.2 KB
    
    📁 before_agent_callbacks/
    
    📁 before_agent_callbacks_01/
    
    📄 python_code.py 5.5 KB
    
    📁 before_model_callbacks/
    
    📁 before_model_callbacks_01/
    
    📄 python_code.py 11.1 KB
    
    📄 instruction.txt 8.7 KB
    
    📄 root_agent.json 1.3 KB
    
    📁 troubleshoot_agent/
    
    📁 before_model_callbacks/
    
    📁 before_model_callbacks_01/
    
    📄 python_code.py 3.5 KB
    
    📄 instruction.txt 6.0 KB
    
    📄 troubleshoot_agent.json 1.4 KB
    
    📁 tools/
    
    📁 lookup_account/
    
    📁 python_function/
    
    📄 python_code.py 4.1 KB
    
    📄 lookup_account.json 1.2 KB
    
    📁 payload_update_tool/
    
    📁 python_function/
    
    📄 python_code.py 2.7 KB
    
    📄 payload_update_tool.json 993 B
    
    📁 set_session_state/
    
    📁 python_function/
    
    📄 python_code.py 2.7 KB
    
    📄 set_session_state.json 1019 B
    
    📄 app.json 5.6 KB
  - 📁 evals/
    
    📁 callback_tests/
    
    📁 agents/
    
    📁 root_agent/
    
    📁 before_model_callbacks/
    
    📁 before_model/
    
    📄 python_code.py 10.8 KB
    
    📁 tests/
    
    📁 root_agent/
    
    📁 before_model_callbacks/
    
    📁 before_model/
    
    📄 test.py 12.0 KB
    
    📁 goldens/
    
    📄 example_auth_routing.yaml 3.9 KB
    
    📄 example_escalation.yaml 2.9 KB
    
    📁 simulations/
    
    📄 example_simulations.yaml 5.6 KB
    
    📁 tool_tests/
    
    📄 example_tool_tests.yaml 2.7 KB
  - 📄 gecx-config.json 983 B
  - 📄 README.md 4.7 KB
  - 📄 tdd.md 4.0 KB
📁 references/
- 📁 api-schemas/
  - 📄 agents.md 1.3 KB
  - 📄 apps.md 1.6 KB
  - 📄 evaluations.md 8.6 KB
  - 📄 sessions.md 2.2 KB
  - 📄 tools.md 1.6 KB
- 📄 api-reference.md 10.8 KB
- 📄 build-verification.md 5.3 KB
- 📄 build.md 14.6 KB
- 📄 callback-api.md 4.0 KB
- 📄 debug.md 17.4 KB
- 📄 eval-templates.md 15.1 KB
- 📄 gecx-design-guide.md 39.2 KB
- 📄 generating-reports.md 11.4 KB
- 📄 interview-guide.md 4.9 KB
- 📄 run.md 5.8 KB
- 📄 setup.md 2.3 KB
📁 scripts/
- 📁 hooks/
  - 📄 post-agent-update.sh 2.6 KB
  - 📄 pre-agent-push-lint.sh 2.6 KB
  - 📄 pre-agent-push.sh 3.8 KB
- 📄 app-thresholds.py 8.3 KB
- 📄 capture-golden-transcripts.py 8.2 KB
- 📄 config.py 5.8 KB
- 📄 configure.py 23.2 KB
- 📄 generate-combined-report.py 49.6 KB
- 📄 generate-iteration-report.py 36.6 KB
- 📄 inspect-app.py 8.9 KB
- 📄 resolve-project.sh 2.4 KB
- 📄 run-all-evals.py 14.5 KB
- 📄 run-and-report.py 4.6 KB
- 📄 scrapi-eval-runner.py 21.1 KB
- 📄 scrapi-sim-runner.py 34.0 KB
- 📄 setup.sh 5.8 KB
- 📄 sync-callbacks.py 7.1 KB
- 📄 triage-results.py 24.0 KB
📄 SKILL.md 3.7 KB

SKILL.md

---
name: cxas-agent-foundry
description: End-to-end GECX/CXAS/CES conversational agent lifecycle -- build agents from requirements (PRD-to-agent), create and run evals (goldens, simulations, tool tests, callback tests), debug failures, and iterate to production quality. Use this skill whenever the user mentions GECX, CXAS, CES, SCRAPI, conversational agents, voice agents, audio agents, agent evals, pushing/pulling/linting agents, or agent instructions/callbacks/tools on the Google Customer Engagement Suite platform.
---

# Agent Foundry

End-to-end lifecycle for GECX conversational agents: build, test, debug, iterate.

## Quick Reference

```bash
# Lint before push (catches structural issues early)
cxas lint --app-dir <project>/cxas_app/<AppName>

# Push local files to platform
cxas push --app-dir <project>/cxas_app/<AppName> \
  --to projects/<project_id>/locations/<location>/apps/<app_id> \
  --project-id <project_id> --location <location>

# Pull platform state to local
cxas pull projects/<project_id>/locations/<location>/apps/<app_id> \
  --project-id <project_id> --location <location> --target-dir <project>/cxas_app/

# Run evals + triage + report (single command)
python .agents/skills/cxas-agent-foundry/scripts/run-and-report.py --message "what changed" --runs 5

# Inspect app architecture
python .agents/skills/cxas-agent-foundry/scripts/inspect-app.py

# Triage failures
python .agents/skills/cxas-agent-foundry/scripts/triage-results.py --last 3
```

## Environment Readiness Check (run BEFORE routing)

Before routing to any sub-skill, check these signals in order:

1. **Virtualenv exists?** -- Check if `.venv/` directory exists
2. **Config exists?** -- Check if `.active-project` file exists and the referenced `<project>/gecx-config.json` exists
3. **Has built before?** -- Check if any `<project>/cxas_app/` directory has content

| Signal | Action |
|--------|--------|
| No `.venv/` or no config | **First-time setup needed.** Load `references/setup.md` before doing anything else. |
| `gecx-config.json` exists but no `cxas_app/` content | Returning user, new project. Route normally. |
| All exist | Returning user. Route normally. |

## Detect Intent and Route

Read what the user wants and load the appropriate sub-skill:

| User says... | Phase | Load |
|-------------|-------|------|
| "Build me an agent from this PRD" | Build | `references/build.md` |
| "Create evals for my agent" | Build | `references/build.md` |
| "Generate tool tests", "create callback tests" | Build | `references/build.md` |
| "Update evals -- requirements changed" | Build | `references/build.md` |
| "Update the TDD" | Build | `references/build.md` |
| "Run evals", "push evals", "check results" | Run | `references/run.md` |
| "Run tool tests", "test the callbacks" | Run | `references/run.md` |
| "Generate a report" | Run | `references/run.md` |
| "Why is this eval failing", "get to 90%" | Debug | `references/debug.md` |
| "Fix the failing evals", "debug the agent" | Debug | `references/debug.md` |
| "Tool test is failing", "callback test broke" | Debug | `references/debug.md` |

If the intent is unclear, ask: "Are you looking to **build/create** evals, **run** them, or **debug** failures?"

## Step Tracking

**CRITICAL:** Before starting any multi-step workflow (Build, Run, or Debug), you MUST create a `todo.md` checklist in the project workspace. 

As you complete each step in the workflow, update the checklist to check it off (`[ ]` to `[x]`). Do not proceed to the next step until the current one is checked off. This ensures no steps are missed during complex builds or debug sessions.

## Before Starting

Check memory for project-specific context (app ID, variable handling rules, audio scoring workarounds). If not available, ask the user.

登录后下载/点赞/收藏 ❤ 19 | ★ 0

cxas-agent-foundry

目录结构

SKILL.md

举报内容

提示