xiexie-qiuligao/counselor.skill
office-call-skill
VerifiedBuild and evolve a single high-fidelity counselor system for real student-counselor communication rehearsal, distillation, correction, and reality-sync practice.
references/case-schema.md
View on GitHub# 案件状态结构
这份文件定义单场景模式下的正式案件结构。
## 目标
让每个 case 都具备一致的状态字段,方便:
- 场景推进
- 结果判定
- 复盘
- 长期档案接入
## 推荐目录结构
```text
generated/cases/{case_id}/
├── case.md
├── state.json
└── transcript.md
```
## state.json 结构
```json
{
"case_id": "20260407-223442-item-48d7d5b1",
"scene_type": "truancy",
"scene_label": "逃课被抓",
"created_at": "2026-04-07T22:34:42",
"updated_at": "2026-04-07T22:34:42",
"stage": "opening",
"status": "active",
"persona_slug": "item-4c1e3222",
"student_profile": {
"display_name": "未命名学生",
"history_flags": [],
"recent_record": []
},
"archive_context": {
"current_focus": [],
"active_flags": [],
"active_threads": [],
"counselor_impression": "neutral",
"trust_trend": "unknown",
"pressure_pattern": [],
"carry_over_notes": []
},
"scores": {
"impression_score": 0,
"trust_level": 0,
"evidence_level": 0,
"risk_level": 0
},
"attitude": {
"current_read": "unknown",
"history": []
},
"evidence_queue": [],
"escalation": {
"current_level": "watch",
"likely_next_steps": [],
"triggers_met": [],
"trigger_log": []
},
"known_info": [],
"hidden_evidence": [],
"redlines": [],
"branch_notes": [],
"result": {
"tier": null,
"label": null,
"reasons": []
},
"next_actions": []
}
```
## 关键字段说明
### `scene_type`
固定类型名,用于决定 playbook:
- `leave-request`
- `truancy`
- `late-return`
### `stage`
推荐阶段:
- `opening`
- `probing`
- `pressure`
- `closure`
- `resolved`
### `scores`
分值并不是游戏 UI,而是内部决策辅助。
建议范围先统一在 `-3` 到 `+3` 或 `0` 到 `5` 内,避免失控。
### `archive_context`
这是深场景最重要的新层。
它的作用是把长期档案中的慢变量带进单场景:
- 当前长期主题
- 累积风险
- 辅导员长期印象
- 当前更习惯的压迫方式
如果没有这层,深场景容易退化成“每局重开”。
### `known_info`
用户已经明确知道的信息。
例如:
- 任课老师已经反馈
- 宿管登记了
- 请假时间冲突
### `hidden_evidence`
不应在开场全部揭示的信息。
例如:
- 班长私下反馈
- 重复记录
- 课程预警线
### `evidence_queue`
推荐把高压场景里的证据做成显式队列,而不是散在文案里。
每条证据至少应包含:
- `label`
- `source`
- `reveal_stage`
- `impact`
- `revealed`
这样才能更稳定地实现:
- 第一轮先试探
- 第二轮亮任课老师反馈
- 第三轮再补班长、宿管或历史记录
### `redlines`
本场景的硬风险触发器。
例如:
- 连续重复发生
- 明显骗过管理
- 事件已经外溢
### `result`
只在结束或接近结束时写定。
### `escalation`
这一层不是处罚单,而是“接下来最可能怎么升级”。
建议记录:
- 当前所处层级
- 最可能的下一步
- 已触发哪些升级条件
- 每次升级是被哪句话或哪条证据触发的
这会直接提高复盘的信息量。
## 最小闭环要求
一个完整 case 至少要做到:
1. 开场有压力
2. 中段有态度判断
3. 至少有一次证据揭示或台阶给予
4. 结尾有明确结果或下一步
5. 长期档案中的至少一个慢变量进入本局
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |