AgentRC 就绪度策略

作者原文

Help the user pick, write, or apply an AgentRC policy. Policies customise readiness scoring by disabling irrelevant checks, overriding impact/level, setting pass-rate thresholds, or chaining org baselines with team overrides. Use when the user asks about strict mode, AI-only scoring, custom weights, CI gating, or wants org-wide standardisation.

SkillPi 解读

能完成的任务

AgentRC 就绪度策略

适合谁

请按作者原文中的适用场景判断。

输入和输出

以作者完整文档为准；本站未运行验证产物。

软件、账号及付费要求

使用前按完整文档确认软件、账号、费用及权限；由宿主核对并授权实际文件读写和外部服务。文档引用域名（非运行白名单）：未列出 HTTP(S) 域名。 调整就绪度评分不等于通过安全审计。禁用检查项或修改CI门槛须解释影响并确认，优先调整权重；npx 从外部获取的实现需固定版本并单独核验。 外部工具按官方来源固定版本并单独核验；本站未安装或运行依赖。文件改动、外部发送、密钥和费用仍由使用者在宿主中逐项授权。

限制与区别

原文中的功能、版本状态和效果描述属于作者陈述；中文为对应翻译。仅核对固定版本及静态附件证据，不保证运行效果、外部服务或兼容性。

实测状态

未测试

安全评级与访问范围

caution · 静态证据核验，不代表运行安全保证。

使用前按完整文档确认软件、账号、费用及权限；由宿主核对并授权实际文件读写和外部服务。文档引用域名（非运行白名单）：未列出 HTTP(S) 域名。 调整就绪度评分不等于通过安全审计。禁用检查项或修改CI门槛须解释影响并确认，优先调整权重；npx 从外部获取的实现需固定版本并单独核验。 外部工具按官方来源固定版本并单独核验；本站未安装或运行依赖。文件改动、外部发送、密钥和费用仍由使用者在宿主中逐项授权。

使用前请由宿主核对并授权实际文件读写、外部服务和联网目的地；本站的核验不会授予任何运行权限。

原始 SKILL.md

---
name: acreadiness-policy
description: 'Help the user pick, write, or apply an AgentRC policy. Policies customise readiness scoring by disabling irrelevant checks, overriding impact/level, setting pass-rate thresholds, or chaining org baselines with team overrides. Use when the user asks about strict mode, AI-only scoring, custom weights, CI gating, or wants org-wide standardisation.'
argument-hint: "[show | new <name> | apply <path-or-pkg>] — e.g. /acreadiness-policy show, /acreadiness-policy new strict-frontend"
---

# /acreadiness-policy — AgentRC policies

Use this skill when the user asks about **policies**, **strict mode**, **custom scoring**, **disabling checks**, **org standards**, or **CI gating** of readiness.

A policy is a small JSON file with three optional sections — `criteria`, `extras`, `thresholds` — that customise how AgentRC scores readiness.

## Built-in examples

AgentRC ships with three example policies in `examples/policies/`:

| Policy | What it does |
|---|---|
| `strict.json` | 100% pass rate, raises impact on key criteria |
| `ai-only.json` | Disables all repo-health checks, focuses on AI tooling |
| `repo-health-only.json` | Disables AI checks, focuses on traditional quality |

Recommend these as starting points before writing a custom policy.

## Policy schema

```jsonc
{
  "name": "my-policy",
  "criteria": {
    "disable":  ["env-example", "observability", "dependabot"],
    "override": {
      "readme":      { "impact": "high", "level": 2 },
      "lint-config": { "title": "Linter required" }
    }
  },
  "extras": {
    "disable": ["pre-commit"]
  },
  "thresholds": {
    "passRate": 0.9
  }
}
```

### Impact weights

| Impact | Weight |
|---|---|
| critical | 5 |
| high | 4 |
| medium | 3 |
| low | 2 |
| info | 0 |

`Score = 1 − (deductions / max possible weight)`. Grades: **A** ≥ 0.9, **B** ≥ 0.8, **C** ≥ 0.7, **D** ≥ 0.6, **F** < 0.6.

## Sub-commands

### `show`
List policies currently in effect (from `agentrc.config.json` `policies` array, or none).

### `new <name>`
Scaffold `policies/<name>.json` with sensible defaults. Walk the user through:
1. **What to disable** — irrelevant pillars or extras for their stack (e.g. disable `observability` for a static site).
2. **What to raise** — override `impact` to `high` or `critical` for must-haves (e.g. `readme`, `codeowners`).
3. **Pass-rate threshold** — typical org baselines: `0.7` (lenient), `0.85` (standard), `1.0` (strict).
4. Reference the policy from `agentrc.config.json`:
   ```json
   { "policies": ["./policies/<name>.json"] }
   ```

### `apply <path-or-pkg>`
Run `agentrc readiness --json --policy <source>` and re-render the report by handing off to the `assess` skill / `ai-readiness-reporter` agent. Supports chaining:
```bash
npx -y github:microsoft/agentrc readiness --json --policy ./org-baseline.json,./team-frontend.json
```

## CI gating

Combine policies with `--fail-level` to enforce a minimum maturity level in CI:

```yaml
- run: npx -y github:microsoft/agentrc readiness --policy ./policies/strict.json --fail-level 3
```

## Advanced

JSON policies can disable, override, and set thresholds — but **cannot add new criteria**. For new detection logic, point users at AgentRC's TypeScript plugin system (`docs/dev/plugins.md`).

## Operating rules

- **Never silently disable a pillar.** If the user wants to disable `observability`, confirm and explain the trade-off.
- **Prefer overriding `impact` over disabling.** Disabling hides the gap entirely; overriding lets it still appear in the report.
- **Recommend extras stay enabled.** They cost nothing — they don't affect the score.
- **Suggest layering** — most orgs want a baseline policy + per-team overrides chained with `--policy a.json,b.json`.

版权与许可声明

MIT License

Copyright GitHub, Inc.

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.