@tea-agent/loop-agent 0.8.0 → 0.9.0
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/AGENTS.md +2 -0
- package/CHANGELOG.md +22 -3
- package/dist/application/dag/args.js +9 -2
- package/dist/executors/shell-executor.js +74 -8
- package/dist/shared/reference-context.js +48 -22
- package/dist/task/config-types.js +1 -1
- package/dist/task/runtime.js +1 -1
- package/dist/worker/loop-agent/loop-agent-client.js +51 -10
- package/dist/worker/observe/routes.js +78 -20
- package/dist/worker/observe/server.js +8 -6
- package/dist/worker/observe/static/app.js +628 -142
- package/dist/worker/observe/static/index.html +68 -43
- package/dist/worker/observe/static/styles.css +478 -602
- package/docs/README.md +4 -0
- package/docs/design/README.md +2 -2
- package/docs/exec-plans/active/README.md +2 -2
- package/docs/exec-plans/completed/README.md +8 -0
- package/docs/reports/README.md +3 -0
- package/package.json +1 -1
- package/scripts/check-task-pool-root.sh +1 -1
- package/skills/loop-agent/references/verification-and-failure-handling.md +8 -3
package/docs/README.md
CHANGED
|
@@ -37,6 +37,10 @@
|
|
|
37
37
|
- `reports/2026-07-11-round3-worker-expansion.md` — Round 3 Worker 扩量、0.7.4 FE green、QA review blocker 与第二轮 owner decision
|
|
38
38
|
- `reports/2026-07-11-observe-dag-visualization.md` — Observe DAG edges、SVG 图形化、交互与真实 Chrome smoke 验证证据
|
|
39
39
|
- `reports/2026-07-11-0.7.5-init-evolution-review.md` — 0.7.5 package/version 变化对目标项目初始化 surface 的影响审查
|
|
40
|
+
- `reports/2026-07-11-command-performance-audit.md` — 命令并发、验证、Observe 和子进程输出的性能审计
|
|
41
|
+
- `reports/2026-07-11-command-performance-followups.md` — Observe 快照复用、输出缓冲和 reference index 遍历优化的验证记录
|
|
42
|
+
- `reports/2026-07-11-observe-terminal-dag-kpi.md` — 终态 DAG 被误计为活跃数的修复记录
|
|
43
|
+
- `reports/2026-07-12-observe-warm-console-redesign.md` — 暖白运行控制台视觉重构与桌面验证记录
|
|
40
44
|
- `decisions/README.md` — 架构决策
|
|
41
45
|
- `skills/README.md` — repo-local skill registry and vetting notes
|
|
42
46
|
- `templates/` — 可复用的规划、报告与 DAG 模板
|
package/docs/design/README.md
CHANGED
|
@@ -8,14 +8,14 @@
|
|
|
8
8
|
|---|---|
|
|
9
9
|
| `产品线共享知识库.md` | 产品线文档仓库作为上游事实源 |
|
|
10
10
|
| `研发模式.md` | 10 个工作日 Feature 团队工作流 |
|
|
11
|
-
| `1个月规划.md` | 首月落地计划 |
|
|
12
|
-
| `1月wbs.md` | 首月 WBS 与分工 |
|
|
13
11
|
| `六个月规划.md` | 六个月路线图与目标架构 |
|
|
14
12
|
| `taskspec-to-loop-agent-mapping.md` | 将产品线 TaskSpec 适配为 `loop-agent` task 的契约 |
|
|
15
13
|
| `state-and-failure-taxonomy.md` | 文档、Task Pool、DAG、Loop 共用的 canonical status 与 failure taxonomy |
|
|
16
14
|
| `observe-ui.md` | 已实现的 agent-worker 本地只读可观测 UI(事件流 / SSE / stale 诊断)设计、事件契约与验收边界 |
|
|
17
15
|
| `observe-ui-optimization.md` | 2026-07-10:**已实现** Observe UI 中文化、DAG-first 总览、进行中 DAG 置顶、read-model enrichment、Pi `session-events.jsonl` 落盘与过程时间线(UI-1~UI-9) |
|
|
18
16
|
| `archive/2026-07-10-下一阶段任务-功能开发完成.md` | 已归档:真实样本、失败硬化与 QA 闭环的功能性开发完成记录 |
|
|
17
|
+
| `archive/2026-07-11-第一月规划.md` | 已归档:首月落地计划(0.8.0 发布 + Round 1/2/3 累计 20 runs / 10 success / 50% / 两次 owner morning decision,第一个月闭环目标全部达成) |
|
|
18
|
+
| `archive/2026-07-11-第一月wbs.md` | 已归档:首月 WBS 与分工(与第一月规划配套,任务编号与估算为历史记录) |
|
|
19
19
|
|
|
20
20
|
## 命名约定(2026-07-09 起)
|
|
21
21
|
|
|
@@ -6,6 +6,6 @@
|
|
|
6
6
|
|
|
7
7
|
当前 active execution plan:
|
|
8
8
|
|
|
9
|
-
- `2026-07-
|
|
9
|
+
- `2026-07-12-observe-warm-console-redesign.md` — 以暖白、细边界和高密度信息架构重构 Observe Dashboard。
|
|
10
10
|
|
|
11
|
-
- 已归档:`../completed/2026-07-11-observe-
|
|
11
|
+
- 已归档:`../completed/2026-07-11-observe-dashboard-page-system.md`、`../completed/2026-07-11-observe-dashboard-detail-refinement.md`、`../completed/2026-07-11-observe-terminal-dag-kpi.md`、`../completed/2026-07-11-observe-polling-efficiency.md`、`../completed/2026-07-11-command-performance-guardrails.md` 及更早计划。
|
|
@@ -4,10 +4,18 @@
|
|
|
4
4
|
|
|
5
5
|
npm 包携带本 README 作为目录契约。具体 completed plan 属于目标仓库历史,不从 loop-agent 源码历史复制。
|
|
6
6
|
|
|
7
|
+
- [`2026-07-11-observe-dashboard-page-system.md`](2026-07-11-observe-dashboard-page-system.md) — 完成 Observe 全页面信息层级与桌面布局优化,按用户确认收口归档。
|
|
8
|
+
- [`2026-07-11-observe-dashboard-detail-refinement.md`](2026-07-11-observe-dashboard-detail-refinement.md) — 完成 DAG 详情输出、依赖图与执行过程的轮询状态保持和桌面节点布局优化,按用户确认收口归档。
|
|
9
|
+
- [`2026-07-11-observe-terminal-dag-kpi.md`](2026-07-11-observe-terminal-dag-kpi.md) — 修复终态 DAG 因遗留未执行节点被误计为活跃的问题。
|
|
10
|
+
- [`2026-07-11-observe-polling-efficiency.md`](2026-07-11-observe-polling-efficiency.md) — 完成详情页终态停轮询、非重叠调度和后台降频。
|
|
11
|
+
- [`2026-07-11-command-performance-guardrails.md`](2026-07-11-command-performance-guardrails.md) — 完成命令并发参数、串行验证与 Observe SSE 的资源峰值护栏。
|
|
12
|
+
- [`2026-07-11-command-performance-followups.md`](2026-07-11-command-performance-followups.md) — 完成 Observe 快照复用、命令输出有界缓冲和 reference index 遍历优化。
|
|
13
|
+
|
|
7
14
|
- [`2026-07-02-loop-agent-subject-restructure.md`](2026-07-02-loop-agent-subject-restructure.md) — 将原 `tools/code-agent` runtime 提升到仓库根、重命名为 `loop-agent`,移除旧 memory plugin 产品线
|
|
8
15
|
- [`2026-07-04-remove-level1-fallback.md`](2026-07-04-remove-level1-fallback.md) — 移除历史顺序 Level 1 fallback,runtime、文档与 command surface 收敛到 DAG 执行
|
|
9
16
|
- [`2026-07-04-runtime-boundary-remediation.md`](2026-07-04-runtime-boundary-remediation.md) — 整合 CLI/skill/runtime 边界,抽出 DAG/Loop runtime seam,集中 harness store/guard 策略
|
|
10
17
|
- [`2026-07-10-next-stage-worker-evidence.md`](2026-07-10-next-stage-worker-evidence.md) — Worker retry、EnvFailure、边界治理、初始化投影与真实 BE/FE/QA dogfood evidence 闭环
|
|
18
|
+
- [`2026-07-10-release-0.6.0-nightly-drill.md`](2026-07-10-release-0.6.0-nightly-drill.md) — 发布 0.6.0、首轮夜间演练(6 worker runs / 3 success)、Observe 过程时间线验证;C6 owner gate 由 Round 3 morning decision 满足后正式关闭
|
|
11
19
|
- [`2026-07-10-harness-json-schema.md`](2026-07-10-harness-json-schema.md) — 新增 `harness.json` IDE JSON Schema、初始化投影、旧项目 update 修复和漂移防护
|
|
12
20
|
- [`2026-07-11-interactive-ui-writer-routing.md`](2026-07-11-interactive-ui-writer-routing.md) — 新增 interactive-ui writer-only HIGH 路由、UI 交付契约、真实 React dogfood fixture 与 Round-2 A/B/C 实验入口
|
|
13
21
|
- [`2026-07-11-round-2-interactive-ui-productization.md`](2026-07-11-round-2-interactive-ui-productization.md) — 完成 owner gate、0.7.0–0.7.4 发布、React A/B/C、20 个 Worker runs、外部 nightly、产品线模板与 docs CI
|
package/docs/reports/README.md
CHANGED
|
@@ -6,4 +6,7 @@
|
|
|
6
6
|
- [`2026-07-10-next-stage-dogfood-runs.md`](2026-07-10-next-stage-dogfood-runs.md) — 发布控制器下的 BE/FE/QA Worker 样本、retry 与 Observe 证据
|
|
7
7
|
- [`2026-07-10-init-evolution-review.md`](2026-07-10-init-evolution-review.md) — Worker 初始化投影演进审查
|
|
8
8
|
- [`2026-07-10-session-events-dogfood.md`](2026-07-10-session-events-dogfood.md) — Pi SDK DAG 新 run 产生非空 `session-events.jsonl` 与 Observe 过程时间线 API 证据
|
|
9
|
+
- [`2026-07-11-command-performance-followups.md`](2026-07-11-command-performance-followups.md) — Observe 快照复用、输出缓冲和 reference index 遍历优化的验证记录
|
|
10
|
+
- [`2026-07-11-observe-terminal-dag-kpi.md`](2026-07-11-observe-terminal-dag-kpi.md) — 终态 DAG 被误计为活跃数的修复记录
|
|
11
|
+
- [`2026-07-12-observe-warm-console-redesign.md`](2026-07-12-observe-warm-console-redesign.md) — 暖白运行控制台视觉重构与桌面验证记录
|
|
9
12
|
- [`2026-07-02-repository-analysis.md`](2026-07-02-repository-analysis.md) — 仓库能力基线分析(2026-07-10 再采样:0.5.0 + Observe UI 优化 UI-1~UI-9 / Unreleased)
|
package/package.json
CHANGED
|
@@ -21,7 +21,7 @@ scan_paths=(
|
|
|
21
21
|
)
|
|
22
22
|
|
|
23
23
|
matches="$(rg -n --path-separator / '[.]task-pool' "${scan_paths[@]}" 2>/dev/null || true)"
|
|
24
|
-
matches="$(printf '%s\n' "${matches}" | rg -v '^(README\.md|AGENTS\.md):.*gitignore|^src/commands/init\.ts:.*("[.]task-pool/"|legacy [.]task-pool residue)|^test/init-command\.test\.ts:.*toContain\("[.]task-pool/"\)' || true)"
|
|
24
|
+
matches="$(printf '%s\n' "${matches}" | rg -v '^(README\.md|AGENTS\.md):.*gitignore|^src/commands/init\.ts:.*("[.]task-pool/"|legacy [.]task-pool residue)|^test/init-command\.test\.ts:.*toContain\("[.]task-pool/"\)|^website/docs/reference/cli\.md:.*旧顶层.*[.]task-pool' || true)"
|
|
25
25
|
|
|
26
26
|
if [[ -n "${matches}" ]]; then
|
|
27
27
|
echo "Task Pool root 检查失败:活跃源码、测试、脚本或用户文档仍引用已废弃的顶层 runtime path:" >&2
|
|
@@ -46,7 +46,7 @@ product-line taxonomy 的事实源是 `docs/design/state-and-failure-taxonomy.md
|
|
|
46
46
|
```json
|
|
47
47
|
{
|
|
48
48
|
"verifyPreset": "auto",
|
|
49
|
-
"verifyMode": "
|
|
49
|
+
"verifyMode": "serial",
|
|
50
50
|
"verifyRetryCount": 0,
|
|
51
51
|
"verifyFailFast": false
|
|
52
52
|
}
|
|
@@ -54,11 +54,16 @@ product-line taxonomy 的事实源是 `docs/design/state-and-failure-taxonomy.md
|
|
|
54
54
|
|
|
55
55
|
- `verifyPreset`:`auto | quick | standard | full`
|
|
56
56
|
- `auto`:`small -> quick`,`medium -> standard`,`large -> full`
|
|
57
|
-
- 默认 `
|
|
58
|
-
-
|
|
57
|
+
- 默认 `serial`,优先避免本机测试、构建和治理检查互相争抢资源
|
|
58
|
+
- 确认命令互不影响且机器资源充足时,才显式使用 `parallel`
|
|
59
59
|
- `verifyFailFast` 仅在 `serial` mode 有意义
|
|
60
60
|
- retry 保持小;用于 flaky 环境问题,不要掩盖真实失败
|
|
61
61
|
|
|
62
|
+
### 长输出处理
|
|
63
|
+
|
|
64
|
+
- DAG shell 命令和 Worker 子进程只在内存中保留最近一段 stdout/stderr,并在结果中标明是否截断及原始字节数。
|
|
65
|
+
- 需要完整日志时读取运行 artifact 中的 `stdout.txt`、`stderr.txt` 或 DAG 节点的 `commands/*.stdout.txt`、`commands/*.stderr.txt`;不要依赖页面摘要还原全部输出。
|
|
66
|
+
|
|
62
67
|
### Verify 后的 goal completion audit
|
|
63
68
|
- 有 active task goal 且 `verify` 通过时,loop-agent 跑 goal completion audit。
|
|
64
69
|
- 若 `source/需求.md` 仍有未勾 checklist(`- [ ]`),task **不会**以 completed 结束。
|