Re-baseline the parity matrix for the full-agent expansion and update the README
Why: the matrix scoped native work to the pi wrapper's exposed surface; the full-agent batch deliberately expands beyond it, so the new domains need their own acceptance basis recorded (standard conformance + negative tests + product acceptance, not pi comparison) plus honest security boundary notes. Evidence: docs-only change; go vet/test unaffected (verified green in the code commits this branch carries). Impact: docs/parity-matrix.md gains the R15 expansion table mapping every domain to its test evidence; README package table reflects the new runtime/mcp, runtime/skills, runtime/promptx, runtime/subagent packages and the v0.2 exec port. Refs: docs/parity-matrix.md #1-#16 remain valid for wrapper-parity rows.
This commit is contained in:
@@ -10,7 +10,7 @@
|
||||
| 层 | 职责 | 依赖边界 |
|
||||
|---|---|---|
|
||||
| contract | 执行契约:Request/Result/Outcome/Error/Capability/Backend/Resolver | 不依赖内核、产品类型、数据库、配置,也不导入错误码库 |
|
||||
| exec | 执行端口:Backend/Request/Result/ProcessEvent、步骤重试与失败分类、交付物物化、Stub、并发限流 | 面向调用方的执行端口 |
|
||||
| exec | 执行端口:Backend/Request/Result/ProcessEvent、预算、步骤重试与失败分类、交付物物化、Stub、并发限流 | 面向调用方的执行端口 |
|
||||
| runtime | 多轮循环、工具调度、模型调用、执行会话 | 不导入 Core/Squad;可选使用日志库,nil 日志器不改变结果 |
|
||||
| adapters | PI Runner transport / typed-result-v1;native:自研内核 runtime → exec.Backend(规范后端名 agent-native,含会话毒化轮转——resume-unsafe 失败后自动换新会话) | 显式映射外部差异,不伪造缺失能力 |
|
||||
| conformance | 所有 Backend 共用的行为验收 | fake 验证契约,真实 adapter 验证能力 |
|
||||
@@ -23,9 +23,13 @@
|
||||
|---|---|
|
||||
| contract | 四态执行契约 |
|
||||
| conformance | 行为验收套件与 fake |
|
||||
| exec | Backend/Request/Result/ProcessEvent、步骤重试与失败分类、交付物物化、Stub、并发限流 |
|
||||
| runtime | 多轮循环、llm、session、tools(含单测) |
|
||||
| adapters/native | runtime 作为 exec.Backend,含会话毒化轮转 |
|
||||
| exec | Backend/Request/Result/ProcessEvent(v0.2:Model/ToolAllowlist/Skills/Budget/SessionScope/ExecutionID/Env)、Budget、步骤重试与失败分类、交付物物化、Stub、并发限流 |
|
||||
| runtime | 多轮循环(预算/取消贯穿、budget_exceeded 事件)、llm、session(compaction/元数据/崩溃安全)、tools(9 内置 + 策略层 Restrict) |
|
||||
| runtime/mcp | MCP 客户端(stdio + Streamable HTTP)与 mcp__<server>__<tool> 桥接(server 元数据不可信、上限收敛) |
|
||||
| runtime/skills | SKILL.md 技能加载与 load_skill 渐进披露 |
|
||||
| runtime/promptx | 系统提示结构化装配(环境事实 + 技能索引 + 交付物约束) |
|
||||
| runtime/subagent | 委派端口(Request/Result/Factory)与 delegate_task 工具;嵌套执行由宿主注入 |
|
||||
| adapters/native | runtime 作为 exec.Backend,含会话毒化轮转、compaction、技能/搜索/执行/MCP/委派装配(Options 显式注入,未配置不注册) |
|
||||
| adapters/pitransport | typed-result-v1 编解码、远程结果解析、Runner 搜索、鉴权头(含 Node 摘要交叉验证测试) |
|
||||
|
||||
本模块不依赖 Core 或 Squad。上层产品依赖本模块的 contract 与 exec。
|
||||
|
||||
@@ -66,3 +66,26 @@ baseline_fixed: true
|
||||
- 本矩阵证据均指向仓库内文件(上文路径),不含运行时截图;
|
||||
- Batch1 完成后在本文件追加"实测证据"列;
|
||||
- 上游仅引固定 0.85.1 的 README/docs 结论,不引 main。
|
||||
|
||||
---
|
||||
|
||||
## R15 扩面批次(full-agent-r1,2026-10-01):从「wrapper 对标」到「完全智能体」
|
||||
|
||||
用户定调(2026-10-01):Agent-Native 补全 pi-Agent 全量工具调用能力(网络搜索、代码编写、
|
||||
系统工具、外部工具调用、技能、提示词),并与 Agent-Squad 融合。扩面域**不做 pi 对标**
|
||||
(上游 wrapper 未暴露这些能力),改以「标准符合 + conformance 负例 + 产品验收」为准——
|
||||
对标口径条款对本批新增域不适用,原有 #1–#16 行继续有效。
|
||||
|
||||
| 域 | 内容 | 出口证据 |
|
||||
|---|---|---|
|
||||
| 执行基座 | exec.Request v0.2(Model/ToolAllowlist/Skills/Budget/SessionScope/ExecutionID/Env)+ Registry.Restrict 策略层(精确名 + 命名空间通配)+ 预算贯穿(token/工具数/截止,budget_exceeded 永久失败)+ 取消传播到工具 | loop_budget / registry_policy / request_v2 / step_retry_budget 测试;关闭 #4/#10/#16 |
|
||||
| 代码工具面 | glob(** 跨目录)、grep 正则/include/ignore_case、read 行范围分页、multi_edit 原子多处替换 | code_tools_test.go |
|
||||
| 系统工具 | run_command / run_python:解释器白名单、env 白名单(宿主密钥不透传)、进程组 SIGKILL(超时/取消)、64KB 头尾截断、命令黑名单、脚本落盘 .agent/scripts/;Options.Exec 未配置不注册 | exec_tools_test.go(含杀组时限断言) |
|
||||
| 网络面 | tavily_search/web_search 经 Options.Search 接线(关闭 #5 尾环);fetch_url:SSRF 硬边界(私网/回环/链路本地/元数据端点逐跳校验)、域名策略、2MB/20s、HTML 剥离 | fetch_test.go + 适配器接线测试 |
|
||||
| MCP | stdio + Streamable HTTP 客户端、initialize/tools/list(paged)/tools/call、Mcp-Session-Id、mcp__<server>__<tool> 桥接、64 工具/1KB 描述/64KB 结果上限、不可达 fail-closed、真实子进程测试(re-exec fake server) | runtime/mcp/mcp_test.go |
|
||||
| 技能/提示词 | SKILL.md 加载(目录名权威、frontmatter 描述)、load_skill 渐进披露、声明技能 fail-closed 校验、promptx 装配器(环境/技能索引/交付物) | skills_test / promptx_test / 适配器测试 |
|
||||
| 会话 | compaction(阈值触发、原子改写、失败留痕不阻断)、元数据 sidecar、崩溃半行容错 | compact_test.go;对齐 #11/#12 |
|
||||
| 委派 | runtime/subagent 端口 + delegate_task(信息集=委派条款);嵌套执行由宿主注入 | subagent_test.go |
|
||||
|
||||
安全声明(不冒充已解决):run_command **不是 OS 沙箱**(宿主负责隔离);技能正文与 MCP
|
||||
server 描述是提示注入面(只信受信根 + 长度上限);fetch 的 DNS 解析存在 rebinding 窗口。
|
||||
|
||||
Reference in New Issue
Block a user