skill-system-gate

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Skill System Gate

Skill 系统门控

Enforce a fail-closed gate before experiment registration and provide read-only registry safety/status checks.

在实验注册前强制执行故障关闭型门控，并提供只读的注册表安全/状态检查。

Decision Matrix

决策矩阵

Task Intent	Operation	Primary Backend	Use When
Validate repository artifacts against configurable gate rules	`validate`	`scripts/gate_validate.py`	Before close-gate decisions or deterministic policy checks
Validate experiment implementation and logs against gate rules	`validate-experiment`	`preprocess_lib.gate_engine`	Legacy GNN/FraudDetect pipeline checks
Check whether registry state is safe for rerun/reset	`verify-registry`	`db_registry.DBExperimentsDB`	Before any manual reset/rerun action
View queue health and state counts	`status`	`db_registry.DBExperimentsDB`	Need a quick dashboard of current experiment states

任务意图	操作	主要后端	适用场景
根据可配置门控规则验证仓库制品	`validate`	`scripts/gate_validate.py`	在做出关闸决策或执行确定性策略检查前
根据门控规则验证实验实现与日志	`validate-experiment`	`preprocess_lib.gate_engine`	旧版GNN/FraudDetect流水线检查
检查注册表状态是否可安全重跑/重置	`verify-registry`	`db_registry.DBExperimentsDB`	在执行任何手动重置/重跑操作前
查看队列健康状态与状态统计	`status`	`db_registry.DBExperimentsDB`	需要快速查看当前实验状态的仪表盘时

Core Patterns

核心模式

Pattern 1: Validate repository gate compliance

模式1：验证仓库门控合规性

bash

python3 scripts/gate_validate.py --rules config/gate_rules.yaml --artifact-root .

Load configurable
```
gate_rules.yaml
```
.
Support rule types:
```
command
```
,
```
structural
```
,
```
eda-contract
```
.
Emit markdown verdict and JSON summary on the last line.

bash

python3 scripts/gate_validate.py --rules config/gate_rules.yaml --artifact-root .

加载可配置的
```
gate_rules.yaml
```
文件。
支持的规则类型：
```
command
```
、
```
structural
```
、
```
eda-contract
```
。
输出markdown格式的判定结果，最后一行是JSON格式的汇总信息。

Pattern 2: Validate legacy experiment gate compliance

模式2：验证旧版实验门控合规性

bash

scripts/validate_exp.sh <exp_name> [phase_root]

Load
```
gate_bank.json
```
rules.

Run all rule classes (

source_contains

source_not_contains

stderr_scan

file_exists

file_min_size

Emit markdown verdict and JSON summary on the last line.

bash

scripts/validate_exp.sh <exp_name> [phase_root]

加载
```
gate_bank.json
```
规则。

运行所有规则类（

source_contains

、

source_not_contains

、

stderr_scan

、

file_exists

、

file_min_size

）。

输出markdown格式的判定结果，最后一行是JSON格式的汇总信息。

Pattern 3: Verify registry safety before reset decisions

模式3：重置决策前验证注册表安全性

bash

scripts/check_registry.sh <exp_name> [phase_root]

Read registry entry for the experiment.
Block reset recommendation for
```
RUNNING
```
experiments.
Return status details and
```
safe_to_reset
```
decision JSON.

bash

scripts/check_registry.sh <exp_name> [phase_root]

读取实验对应的注册表条目。
对处于
```
RUNNING
```
状态的实验，禁止重置操作建议。
返回状态详情与
```
safe_to_reset
```
决策的JSON数据。

Pattern 4: Show current registry health dashboard

模式4：查看当前注册表健康仪表盘

bash

scripts/status.sh [phase_root]

Aggregate counts across active/completed sets.
Show running experiments and execution location.
Return compact JSON counters on the last line.

bash

scripts/status.sh [phase_root]

汇总活跃/已完成实验的统计数量。
展示运行中的实验及其执行位置。
最后一行返回紧凑的JSON格式统计计数器。

Guidelines

指导原则

Always treat validation as fail-closed: any error-severity rule violation means gate failure.
Always run
```
validate-experiment
```
before registry-facing actions.
Never modify experiment source/config files during checks.
Never execute reset/rerun from this skill; only report recommendations.
Never override
```
RUNNING
```
experiments as safe to reset.
Keep all operations read-only with respect to registry and experiment artifacts.

skill

{
  "schema_version": "2.0",
  "id": "skill-system-gate",
  "version": "1.0.0",
  "capabilities": ["configurable-gate", "experiment-gate", "registry-safety", "experiment-status"],
  "effects": ["fs.read", "db.read", "proc.exec"],
  "operations": {
    "validate-experiment": {
      "description": "Validate experiment source artifacts and logs against gate_bank rules and return a pass/fail verdict.",
      "input": {
        "exp_name": { "type": "string", "required": true, "description": "Experiment directory name under experiments/" },
        "phase_root": { "type": "string", "required": false, "description": "Phase directory path", "default": "/datas/store162/arthur0824hao/Study/GNN/FraudDetect/SubProject/Phase3" }
      },
      "output": {
        "description": "Gate validation markdown plus machine-readable verdict.",
        "fields": { "passed": "boolean", "errors": "number", "warnings": "number" }
      },
      "entrypoints": {
        "unix": ["bash", "scripts/validate_exp.sh", "{exp_name}", "{phase_root}"]
      }
    },
    "validate": {
      "description": "Run configurable gate_rules.yaml checks and emit pass/fail with per-rule results.",
      "input": {
        "rules": { "type": "string", "required": false, "description": "Path to gate rules file", "default": "config/gate_rules.yaml" },
        "artifact_root": { "type": "string", "required": false, "description": "Working directory for rule execution", "default": "." }
      },
      "output": {
        "description": "Gate verdict with per-rule details.",
        "fields": { "passed": "boolean", "errors": "number", "warnings": "number" }
      },
      "entrypoints": {
        "unix": ["python3", "scripts/gate_validate.py", "--rules", "{rules}", "--artifact-root", "{artifact_root}"]
      }
    },
    "verify-registry": {
      "description": "Check registry state safety and recommend whether reset/rerun is safe without mutating state.",
      "input": {
        "exp_name": { "type": "string", "required": true, "description": "Experiment name in registry" },
        "phase_root": { "type": "string", "required": false, "description": "Phase directory path", "default": "/datas/store162/arthur0824hao/Study/GNN/FraudDetect/SubProject/Phase3" }
      },
      "output": {
        "description": "Registry safety markdown plus recommendation payload.",
        "fields": { "found": "boolean", "status": "string", "safe_to_reset": "boolean" }
      },
      "entrypoints": {
        "unix": ["bash", "scripts/check_registry.sh", "{exp_name}", "{phase_root}"]
      }
    },
    "status": {
      "description": "Show current registry dashboard with counts by status and running experiment details.",
      "input": {
        "phase_root": { "type": "string", "required": false, "description": "Phase directory path", "default": "/datas/store162/arthur0824hao/Study/GNN/FraudDetect/SubProject/Phase3" }
      },
      "output": {
        "description": "Registry status dashboard and aggregate counters.",
        "fields": { "running": "number", "needs_rerun": "number", "completed": "number", "total": "number" }
      },
      "entrypoints": {
        "unix": ["bash", "scripts/status.sh", "{phase_root}"]
      }
    }
  },
  "stdout_contract": {
    "last_line_json": true
  }
}

始终将验证视为故障关闭型：任何错误级别的规则违反都意味着门控失败。
在执行任何面向注册表的操作前，务必运行
```
validate-experiment
```
。
检查过程中切勿修改实验源码/配置文件。
切勿通过该Skill执行重置/重跑操作；仅可输出操作建议。
切勿将处于
```
RUNNING
```
状态的实验标记为可安全重置。
所有操作对注册表与实验制品均保持只读权限。

skill

{
  "schema_version": "2.0",
  "id": "skill-system-gate",
  "version": "1.0.0",
  "capabilities": ["configurable-gate", "experiment-gate", "registry-safety", "experiment-status"],
  "effects": ["fs.read", "db.read", "proc.exec"],
  "operations": {
    "validate-experiment": {
      "description": "Validate experiment source artifacts and logs against gate_bank rules and return a pass/fail verdict.",
      "input": {
        "exp_name": { "type": "string", "required": true, "description": "Experiment directory name under experiments/" },
        "phase_root": { "type": "string", "required": false, "description": "Phase directory path", "default": "/datas/store162/arthur0824hao/Study/GNN/FraudDetect/SubProject/Phase3" }
      },
      "output": {
        "description": "Gate validation markdown plus machine-readable verdict.",
        "fields": { "passed": "boolean", "errors": "number", "warnings": "number" }
      },
      "entrypoints": {
        "unix": ["bash", "scripts/validate_exp.sh", "{exp_name}", "{phase_root}"]
      }
    },
    "validate": {
      "description": "Run configurable gate_rules.yaml checks and emit pass/fail with per-rule results.",
      "input": {
        "rules": { "type": "string", "required": false, "description": "Path to gate rules file", "default": "config/gate_rules.yaml" },
        "artifact_root": { "type": "string", "required": false, "description": "Working directory for rule execution", "default": "." }
      },
      "output": {
        "description": "Gate verdict with per-rule details.",
        "fields": { "passed": "boolean", "errors": "number", "warnings": "number" }
      },
      "entrypoints": {
        "unix": ["python3", "scripts/gate_validate.py", "--rules", "{rules}", "--artifact-root", "{artifact_root}"]
      }
    },
    "verify-registry": {
      "description": "Check registry state safety and recommend whether reset/rerun is safe without mutating state.",
      "input": {
        "exp_name": { "type": "string", "required": true, "description": "Experiment name in registry" },
        "phase_root": { "type": "string", "required": false, "description": "Phase directory path", "default": "/datas/store162/arthur0824hao/Study/GNN/FraudDetect/SubProject/Phase3" }
      },
      "output": {
        "description": "Registry safety markdown plus recommendation payload.",
        "fields": { "found": "boolean", "status": "string", "safe_to_reset": "boolean" }
      },
      "entrypoints": {
        "unix": ["bash", "scripts/check_registry.sh", "{exp_name}", "{phase_root}"]
      }
    },
    "status": {
      "description": "Show current registry dashboard with counts by status and running experiment details.",
      "input": {
        "phase_root": { "type": "string", "required": false, "description": "Phase directory path", "default": "/datas/store162/arthur0824hao/Study/GNN/FraudDetect/SubProject/Phase3" }
      },
      "output": {
        "description": "Registry status dashboard and aggregate counters.",
        "fields": { "running": "number", "needs_rerun": "number", "completed": "number", "total": "number" }
      },
      "entrypoints": {
        "unix": ["bash", "scripts/status.sh", "{phase_root}"]
      }
    }
  },
  "stdout_contract": {
    "last_line_json": true
  }
}