mirror of
https://bitbucket.org/siakitem/my-pi.git
synced 2026-10-26 18:55:22 +00:00
云服务部-叶林立andDevin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
7e6a8faae1
AutoReview： - 审查器回到 Codex Guardian 风格的对话式审查，默认跟随会话当前主模型 （每次审查实时读取 ctx.model）；provider/model 必须同时设置才固定审查 模型，只设其一视为配置错误并转人工。 - 结论只通过 approve / reject 两个工具给出，参数仅 reason；不强制 toolChoice，正文文本（即使是合法 JSON）一律忽略。 - approve 自动放行；reject 转人工并在弹窗显示理由（标注模型生成）。 超时、Provider 错误、截断、未调用工具、多次调用、参数无效等异常同样 转人工，并附固定的系统说明；用户按 Esc 中止时直接拒绝。 - 模型调用改为 modelRegistry.streamSimple(...).result()：Pi 1.3.0 面向 扩展的 ModelRegistry 没有 completeSimple，原写法在真实环境中每次都会 转人工。 - 对话记录保留可信度标注；工具结果在有更短的 pi-compaction 摘要时替换 为摘要；token 估算改为非 ASCII 字符按 1 token 计；预算可配置。 - 同一条用户消息内完全相同的请求只缓存 approve 结论。 - 默认 reasoning low、总超时 30 秒、最多 2 次尝试、maxOutputTokens 4000。 删除： - 未发布过的 TypeSafe Jev 分类器实现、DenialCircuitBreaker。 - 整个 annotator 机制（model-explanation、注册表、registerAnnotator、 provider-timeout）；annotators 配置键保留为已弃用并忽略，避免旧配置 被拒绝。 测试： - 新增 reviewer / transcript / 弹窗说明 / 默认配置测试；清理重复、恒真 或只测实现细节的旧测试。 - 使用 deepseek-v4.1-flash 实测 5 个场景各 5 次，25 次结论全部符合预期。 Generated with [Devin](https://devin.ai) Co-Authored-By: Devin <158243242+devin-ai-integration[bot]@users.noreply.github.com>
95 lines
3.0 KiB
TypeScript
95 lines
3.0 KiB
TypeScript
import { describe, expect, it } from "vitest";
|
|
|
|
import { describeSkillInputGate } from "#src/handlers/gates/skill-input";
|
|
import { makeCheckResult } from "#test/helpers/handler-fixtures";
|
|
|
|
// ── helpers ────────────────────────────────────────────────────────────────
|
|
|
|
function makeSkillCheck(state: "allow" | "deny" | "ask") {
|
|
return makeCheckResult({
|
|
state,
|
|
toolName: "skill",
|
|
source: "skill",
|
|
origin: "global",
|
|
matchedPattern: "*",
|
|
});
|
|
}
|
|
|
|
// ── describeSkillInputGate ─────────────────────────────────────────────────
|
|
|
|
describe("describeSkillInputGate", () => {
|
|
it("makes the skill the payload's decision-relevant value", () => {
|
|
const descriptor = describeSkillInputGate(
|
|
"librarian",
|
|
null,
|
|
makeSkillCheck("allow"),
|
|
);
|
|
expect(descriptor.surface).toBe("skill");
|
|
expect(descriptor.input).toEqual({ name: "librarian" });
|
|
expect(descriptor.payload.kind).toBe("skill");
|
|
expect(descriptor.payload.request.surface).toBe("skill");
|
|
expect(descriptor.payload.request.value).toBe("librarian");
|
|
expect(descriptor.payload.request.requester.agentName).toBeNull();
|
|
});
|
|
|
|
it("names the requesting agent on the payload when provided", () => {
|
|
const descriptor = describeSkillInputGate(
|
|
"librarian",
|
|
"code-agent",
|
|
makeSkillCheck("allow"),
|
|
);
|
|
expect(descriptor.payload.request.requester.agentName).toBe("code-agent");
|
|
});
|
|
|
|
it("sets promptDetails source to 'skill_input' with skill name and agent", () => {
|
|
const descriptor = describeSkillInputGate(
|
|
"librarian",
|
|
"code-agent",
|
|
makeSkillCheck("ask"),
|
|
);
|
|
expect(descriptor.promptDetails).toMatchObject({
|
|
source: "skill_input",
|
|
agentName: "code-agent",
|
|
skillName: "librarian",
|
|
});
|
|
});
|
|
|
|
it("sets logContext source to 'skill_input' with skill name and agent", () => {
|
|
const descriptor = describeSkillInputGate(
|
|
"librarian",
|
|
"code-agent",
|
|
makeSkillCheck("allow"),
|
|
);
|
|
expect(descriptor.logContext).toMatchObject({
|
|
source: "skill_input",
|
|
skillName: "librarian",
|
|
agentName: "code-agent",
|
|
});
|
|
});
|
|
|
|
it("sets decision surface to 'skill' and value to the skill name", () => {
|
|
const descriptor = describeSkillInputGate(
|
|
"my-skill",
|
|
null,
|
|
makeSkillCheck("allow"),
|
|
);
|
|
expect(descriptor.decision).toEqual({
|
|
surface: "skill",
|
|
value: "my-skill",
|
|
});
|
|
});
|
|
|
|
it("carries the skill name as single-value access facts on promptDetails", () => {
|
|
const descriptor = describeSkillInputGate(
|
|
"my-skill",
|
|
null,
|
|
makeSkillCheck("ask"),
|
|
);
|
|
expect(descriptor.promptDetails.accessIntent).toEqual({
|
|
surface: "skill",
|
|
matchValues: ["my-skill"],
|
|
boundaryValue: null,
|
|
});
|
|
});
|
|
});
|