@tea-agent/loop-agent 0.4.0 → 0.5.0

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.
Files changed (106) hide show
  1. package/AGENTS.md +142 -142
  2. package/CHANGELOG.md +98 -106
  3. package/README.md +195 -195
  4. package/bin/agent-worker.js +22 -22
  5. package/bin/loop-agent.js +21 -21
  6. package/dist/application/dag/validate-dag.js +14 -1
  7. package/dist/commands/init.js +677 -489
  8. package/dist/commands/loop-benchmark.js +11 -11
  9. package/dist/commands/pi-reuse-benchmark.js +16 -16
  10. package/dist/executors/config-core.js +3 -2
  11. package/dist/executors/cursor-executor.js +1 -1
  12. package/dist/executors/model-routing.js +43 -0
  13. package/dist/governance/manifest-types.js +9 -1
  14. package/dist/task/runtime.js +27 -27
  15. package/dist/worker/pool/run-store.js +9 -1
  16. package/dist/workflows/dag/canvas-observer.js +275 -275
  17. package/dist/workflows/dag/init-hybrid.js +4 -13
  18. package/dist/workflows/dag/skill-instructions.js +4 -0
  19. package/dist/workflows/dag/types.js +1 -1
  20. package/dist/workflows/dag/validate.js +3 -2
  21. package/docs/README.md +72 -72
  22. package/docs/agent-dag-recovery-playbook.md +184 -184
  23. package/docs/agent-dag-runner.md +42 -42
  24. package/docs/architecture/runtime-boundaries.md +147 -147
  25. package/docs/cursor-executor-usage.md +25 -25
  26. package/docs/decisions/README.md +3 -3
  27. package/docs/design/README.md +36 -36
  28. package/docs/development-principles.md +73 -73
  29. package/docs/dynamic-workflow-dag-engine-roadmap.md +1749 -1749
  30. package/docs/exec-plans/README.md +6 -6
  31. package/docs/exec-plans/active/README.md +7 -7
  32. package/docs/exec-plans/completed/README.md +19 -19
  33. package/docs/feature-workflow.md +186 -186
  34. package/docs/harness-methodology-debugging.md +153 -153
  35. package/docs/harness-methodology-tdd.md +130 -130
  36. package/docs/harness-methodology-verification.md +27 -27
  37. package/docs/init-surface.manifest.json +199 -175
  38. package/docs/loop-agent-harness.md +42 -42
  39. package/docs/production-readiness.md +96 -96
  40. package/docs/progress/README.md +3 -3
  41. package/docs/reports/README.md +5 -5
  42. package/docs/skills/README.md +6 -6
  43. package/docs/skills/vetted-skill-registry.md +26 -26
  44. package/docs/templates/adr.md +60 -60
  45. package/docs/templates/agent-dag-authority-surface-audit.prompt.md +94 -94
  46. package/docs/templates/agent-dag-decision-envelope.schema.json +213 -213
  47. package/docs/templates/agent-dag-decision-gate-dogfood-report.md +117 -117
  48. package/docs/templates/agent-dag-decision-gate.prompt.md +246 -246
  49. package/docs/templates/agent-dag-process-supervisor.prompt.md +98 -98
  50. package/docs/templates/agent-dag-report.schema.json +454 -454
  51. package/docs/templates/agent-dag-review-verdict.prompt.md +68 -68
  52. package/docs/templates/agent-dag.base.json +195 -195
  53. package/docs/templates/agent-dag.final-verification.json +190 -190
  54. package/docs/templates/agent-dag.schema.json +316 -316
  55. package/docs/templates/agent-dag.supervised-implementation.json +500 -500
  56. package/docs/templates/exec-plan.md +64 -64
  57. package/docs/templates/feature-spec.md +53 -53
  58. package/docs/templates/hybrid-dag.json +193 -193
  59. package/docs/templates/init-evolution-review.md +33 -33
  60. package/docs/templates/production-readiness-checklist.md +57 -57
  61. package/docs/templates/progress-log.md +17 -17
  62. package/docs/templates/project-start-checklist.md +9 -9
  63. package/docs/templates/qa-report.md +48 -48
  64. package/docs/templates/sprint-contract.md +29 -29
  65. package/docs/verification-matrix.md +41 -41
  66. package/examples/decision-gate-agent-dag.json +123 -123
  67. package/examples/example-dag.json +51 -51
  68. package/examples/hybrid-loop-agent-dag.json +194 -194
  69. package/harness.json +69 -94
  70. package/package.json +66 -66
  71. package/skills/ai-engineering-context/SKILL.md +48 -48
  72. package/skills/code-review-core/SKILL.md +20 -20
  73. package/skills/codebase-scout/SKILL.md +19 -19
  74. package/skills/init-capability-evolution/SKILL.md +69 -69
  75. package/skills/loop-agent/SKILL.md +147 -147
  76. package/skills/loop-agent/references/README.md +67 -67
  77. package/skills/loop-agent/references/command-reference.md +403 -403
  78. package/skills/loop-agent/references/harness-policy.md +259 -259
  79. package/skills/loop-agent/references/hybrid-dag.md +216 -216
  80. package/skills/loop-agent/references/learned/README.md +21 -21
  81. package/skills/loop-agent/references/long-running-loop.md +59 -59
  82. package/skills/loop-agent/references/model-routing.md +36 -36
  83. package/skills/loop-agent/references/multi-worktree.md +54 -54
  84. package/skills/loop-agent/references/one-shot-runs.md +85 -85
  85. package/skills/loop-agent/references/orchestrator-and-interventions.md +169 -169
  86. package/skills/loop-agent/references/pi-prompt.md +23 -23
  87. package/skills/loop-agent/references/pi-subagent-assisted-mode.md +81 -81
  88. package/skills/loop-agent/references/post-implementation-and-patterns.md +44 -44
  89. package/skills/loop-agent/references/task-workflow.md +84 -84
  90. package/skills/loop-agent/references/verification-and-failure-handling.md +128 -128
  91. package/skills/requesting-code-review/SKILL.md +101 -101
  92. package/skills/requesting-code-review/code-reviewer.md +168 -168
  93. package/skills/systematic-debugging/CREATION-LOG.md +119 -119
  94. package/skills/systematic-debugging/SKILL.md +296 -296
  95. package/skills/systematic-debugging/condition-based-waiting-example.ts +158 -158
  96. package/skills/systematic-debugging/condition-based-waiting.md +115 -115
  97. package/skills/systematic-debugging/defense-in-depth.md +122 -122
  98. package/skills/systematic-debugging/find-polluter.sh +63 -63
  99. package/skills/systematic-debugging/root-cause-tracing.md +169 -169
  100. package/skills/systematic-debugging/test-academic.md +14 -14
  101. package/skills/systematic-debugging/test-pressure-1.md +58 -58
  102. package/skills/systematic-debugging/test-pressure-2.md +68 -68
  103. package/skills/systematic-debugging/test-pressure-3.md +69 -69
  104. package/skills/test-driven-development/SKILL.md +20 -20
  105. package/skills/verification-before-completion/SKILL.md +154 -154
  106. package/skills/webapp-testing/SKILL.md +19 -19
@@ -1,33 +1,33 @@
1
- # Init Evolution Review
2
-
3
- Date:
4
- Base:
5
- Head:
6
-
7
- ## Changed Surface
8
-
9
- -
10
-
11
- ## Decision
12
-
13
- Choose one:
14
-
15
- - No init impact
16
- - Surface check only
17
- - Init update required
18
-
19
- Rationale:
20
-
21
- ## Updates Made
22
-
23
- -
24
-
25
- ## Verification
26
-
27
- ```bash
28
- # commands and results
29
- ```
30
-
31
- ## Residual Risk
32
-
33
- -
1
+ # Init Evolution Review
2
+
3
+ Date:
4
+ Base:
5
+ Head:
6
+
7
+ ## Changed Surface
8
+
9
+ -
10
+
11
+ ## Decision
12
+
13
+ Choose one:
14
+
15
+ - No init impact
16
+ - Surface check only
17
+ - Init update required
18
+
19
+ Rationale:
20
+
21
+ ## Updates Made
22
+
23
+ -
24
+
25
+ ## Verification
26
+
27
+ ```bash
28
+ # commands and results
29
+ ```
30
+
31
+ ## Residual Risk
32
+
33
+ -
@@ -1,57 +1,57 @@
1
- # Production Readiness 检查清单
2
-
3
- 声称 Production Readiness v0.1 的低/中风险单仓库 DAG 任务使用本清单。
4
-
5
- ## 范围
6
-
7
- - [ ] 单仓库
8
- - [ ] 单任务或小范围有边界任务
9
- - [ ] 低或中风险
10
- - [ ] 无生产 secret 或生产数据库访问
11
- - [ ] 无自动 merge 或 release
12
- - [ ] 无 DAG runtime 之外的第二套 runner
13
- - [ ] 无可写 Dynamic Workflow sharded migration
14
-
15
- ## Task Contract
16
-
17
- - [ ] Task source 存在于 `.harness/tasks/<task-id>/source/`
18
- - [ ] `allowedPaths` 显式
19
- - [ ] `forbiddenPaths` 显式
20
- - [ ] `writeSet` 或预期写 scope 显式
21
- - [ ] Shell 验证命令显式
22
- - [ ] 非目标显式
23
-
24
- ## DAG 主路径
25
-
26
- - [ ] 已运行 `loop-agent new-task <task-id> "Task title"` 或 task 已存在
27
- - [ ] `loop-agent dag run-task <task-id> --profile auto --strict-models --output <dag-path>` 产出 DAG spec
28
- - [ ] `loop-agent dag validate --dag <dag-path> --strict-models --strict-governance` 通过
29
- - [ ] `loop-agent run-dag --dag <dag-path> --cwd .` 产出 run id
30
- - [ ] `loop-agent dag report --run-id <run-id> --markdown` 可读
31
- - [ ] `loop-agent dag doctor --run-id <run-id>` 能解释失败或 paused run
32
-
33
- ## 证据
34
-
35
- - [ ] 记录 DAG spec path
36
- - [ ] 记录 DAG validation 输出
37
- - [ ] 记录 run id
38
- - [ ] Shell 验证输出是新鲜的
39
- - [ ] 成功路径有 promotion 与 closeout 证据
40
- - [ ] 失败路径有 failure handoff 证据
41
-
42
- ## Failure Routing
43
-
44
- - [ ] 存在时保留 raw failure category
45
- - [ ] 失败 run 有 DAG normalized failure category
46
- - [ ] 失败 run 有 product-line failure category
47
- - [ ] 失败 run 有 recommended follow-up
48
- - [ ] 派生 category 未覆盖已完成 DAG facts
49
- - [ ] `Unknown` 已说明或为 fixture 有意接受
50
-
51
- ## 最终门禁
52
-
53
- - [ ] `npm run typecheck`
54
- - [ ] 相关定向 Vitest 文件
55
- - [ ] `bash scripts/check-repo.sh`
56
- - [ ] website/docs 变更时 `npm run docs:build`
57
- - [ ] 最终 sprint closeout 前 `bash scripts/ci.sh`
1
+ # Production Readiness 检查清单
2
+
3
+ 声称 Production Readiness v0.1 的低/中风险单仓库 DAG 任务使用本清单。
4
+
5
+ ## 范围
6
+
7
+ - [ ] 单仓库
8
+ - [ ] 单任务或小范围有边界任务
9
+ - [ ] 低或中风险
10
+ - [ ] 无生产 secret 或生产数据库访问
11
+ - [ ] 无自动 merge 或 release
12
+ - [ ] 无 DAG runtime 之外的第二套 runner
13
+ - [ ] 无可写 Dynamic Workflow sharded migration
14
+
15
+ ## Task Contract
16
+
17
+ - [ ] Task source 存在于 `.harness/tasks/<task-id>/source/`
18
+ - [ ] `allowedPaths` 显式
19
+ - [ ] `forbiddenPaths` 显式
20
+ - [ ] `writeSet` 或预期写 scope 显式
21
+ - [ ] Shell 验证命令显式
22
+ - [ ] 非目标显式
23
+
24
+ ## DAG 主路径
25
+
26
+ - [ ] 已运行 `loop-agent new-task <task-id> "Task title"` 或 task 已存在
27
+ - [ ] `loop-agent dag run-task <task-id> --profile auto --strict-models --output <dag-path>` 产出 DAG spec
28
+ - [ ] `loop-agent dag validate --dag <dag-path> --strict-models --strict-governance` 通过
29
+ - [ ] `loop-agent run-dag --dag <dag-path> --cwd .` 产出 run id
30
+ - [ ] `loop-agent dag report --run-id <run-id> --markdown` 可读
31
+ - [ ] `loop-agent dag doctor --run-id <run-id>` 能解释失败或 paused run
32
+
33
+ ## 证据
34
+
35
+ - [ ] 记录 DAG spec path
36
+ - [ ] 记录 DAG validation 输出
37
+ - [ ] 记录 run id
38
+ - [ ] Shell 验证输出是新鲜的
39
+ - [ ] 成功路径有 promotion 与 closeout 证据
40
+ - [ ] 失败路径有 failure handoff 证据
41
+
42
+ ## Failure Routing
43
+
44
+ - [ ] 存在时保留 raw failure category
45
+ - [ ] 失败 run 有 DAG normalized failure category
46
+ - [ ] 失败 run 有 product-line failure category
47
+ - [ ] 失败 run 有 recommended follow-up
48
+ - [ ] 派生 category 未覆盖已完成 DAG facts
49
+ - [ ] `Unknown` 已说明或为 fixture 有意接受
50
+
51
+ ## 最终门禁
52
+
53
+ - [ ] `npm run typecheck`
54
+ - [ ] 相关定向 Vitest 文件
55
+ - [ ] `bash scripts/check-repo.sh`
56
+ - [ ] website/docs 变更时 `npm run docs:build`
57
+ - [ ] 最终 sprint closeout 前 `bash scripts/ci.sh`
@@ -1,17 +1,17 @@
1
- # 进度日志模板
2
-
3
- ## 日期 / 会话
4
-
5
- ## 变更内容
6
-
7
- ## 已验证项
8
-
9
- ## 新发现的 Bug / 技术债 / 漂移
10
-
11
- - Bugs:
12
- - Debt:
13
- - Drift:
14
-
15
- ## 已知缺口 / 风险
16
-
17
- ## 推荐下一步
1
+ # 进度日志模板
2
+
3
+ ## 日期 / 会话
4
+
5
+ ## 变更内容
6
+
7
+ ## 已验证项
8
+
9
+ ## 新发现的 Bug / 技术债 / 漂移
10
+
11
+ - Bugs:
12
+ - Debt:
13
+ - Drift:
14
+
15
+ ## 已知缺口 / 风险
16
+
17
+ ## 推荐下一步
@@ -1,9 +1,9 @@
1
- # 项目开工检查清单
2
-
3
- - [ ] 确认 `pwd`
4
- - [ ] 阅读 `README.md`、`harness.json`、`docs/README.md`
5
- - [ ] 检查 `git status --short --branch`
6
- - [ ] 确定单一工作块
7
- - [ ] 从 `docs/verification-matrix.md` 选择验证命令
8
- - [ ] 保留无关用户变更
9
- - [ ] 非平凡工作时记录 handoff 证据
1
+ # 项目开工检查清单
2
+
3
+ - [ ] 确认 `pwd`
4
+ - [ ] 阅读 `README.md`、`harness.json`、`docs/README.md`
5
+ - [ ] 检查 `git status --short --branch`
6
+ - [ ] 确定单一工作块
7
+ - [ ] 从 `docs/verification-matrix.md` 选择验证命令
8
+ - [ ] 保留无关用户变更
9
+ - [ ] 非平凡工作时记录 handoff 证据
@@ -1,48 +1,48 @@
1
- # QA 报告模板
2
-
3
- ## 日期 / 会话
4
-
5
- ## 测试范围
6
-
7
- ## 相关 Contract / Plan
8
-
9
- ## 测试方法
10
-
11
- - 自动化测试:
12
- - 构建 / 类型检查:
13
- - Smoke / 手工路径:
14
- - 其他证据:
15
-
16
- ## Smoke / 手工检查
17
-
18
- - CLI smoke:
19
- - DAG smoke:
20
- - Loop workflow smoke:
21
- - Executor smoke:
22
- - 结果:pass / blocked / fail
23
-
24
- ## 通过标准
25
-
26
- ## 发现项
27
-
28
- | ID | Severity | Finding | Evidence | Suggested Fix |
29
- |----|----------|---------|----------|---------------|
30
-
31
- ## Failure Routing
32
-
33
- | Run / Scenario | Raw Failure | DAG Normalized Category | Product-line Category | Recommended Follow-up | Evidence |
34
- |---|---|---|---|---|---|
35
- | | | | | | |
36
-
37
- ## 新发现的 Bug / 漂移
38
-
39
- - 新发现 bug:
40
- - 发现的契约漂移:
41
- - 发现的文档/测试不一致:
42
-
43
- ## 最终结论
44
-
45
- - 结论:pass / partial / fail
46
- - 结论依据:
47
-
48
- ## 后续项
1
+ # QA 报告模板
2
+
3
+ ## 日期 / 会话
4
+
5
+ ## 测试范围
6
+
7
+ ## 相关 Contract / Plan
8
+
9
+ ## 测试方法
10
+
11
+ - 自动化测试:
12
+ - 构建 / 类型检查:
13
+ - Smoke / 手工路径:
14
+ - 其他证据:
15
+
16
+ ## Smoke / 手工检查
17
+
18
+ - CLI smoke:
19
+ - DAG smoke:
20
+ - Loop workflow smoke:
21
+ - Executor smoke:
22
+ - 结果:pass / blocked / fail
23
+
24
+ ## 通过标准
25
+
26
+ ## 发现项
27
+
28
+ | ID | Severity | Finding | Evidence | Suggested Fix |
29
+ |----|----------|---------|----------|---------------|
30
+
31
+ ## Failure Routing
32
+
33
+ | Run / Scenario | Raw Failure | DAG Normalized Category | Product-line Category | Recommended Follow-up | Evidence |
34
+ |---|---|---|---|---|---|
35
+ | | | | | | |
36
+
37
+ ## 新发现的 Bug / 漂移
38
+
39
+ - 新发现 bug:
40
+ - 发现的契约漂移:
41
+ - 发现的文档/测试不一致:
42
+
43
+ ## 最终结论
44
+
45
+ - 结论:pass / partial / fail
46
+ - 结论依据:
47
+
48
+ ## 后续项
@@ -1,29 +1,29 @@
1
- # Sprint Contract
2
-
3
- ## 目标(Objective)
4
-
5
- 描述有边界的工作块。
6
-
7
- ## 交付物(Deliverables)
8
-
9
- -
10
-
11
- ## 非目标(Non-goals)
12
-
13
- -
14
-
15
- ## 验证(Verification)
16
-
17
- ```bash
18
- bash scripts/check-repo.sh
19
- npm run typecheck
20
- npm test
21
- ```
22
-
23
- Windows 上通过 Git Bash 或已配置的兼容 Bash 运行脚本。实际文件操作用平台原生路径。
24
-
25
- ## 失败条件(Failure Conditions)
26
-
27
- - 必需验证无法运行或失败
28
- - 工作需要超出本 contract 的范围
29
- - 实现改动了无关文件
1
+ # Sprint Contract
2
+
3
+ ## 目标(Objective)
4
+
5
+ 描述有边界的工作块。
6
+
7
+ ## 交付物(Deliverables)
8
+
9
+ -
10
+
11
+ ## 非目标(Non-goals)
12
+
13
+ -
14
+
15
+ ## 验证(Verification)
16
+
17
+ ```bash
18
+ bash scripts/check-repo.sh
19
+ npm run typecheck
20
+ npm test
21
+ ```
22
+
23
+ Windows 上通过 Git Bash 或已配置的兼容 Bash 运行脚本。实际文件操作用平台原生路径。
24
+
25
+ ## 失败条件(Failure Conditions)
26
+
27
+ - 必需验证无法运行或失败
28
+ - 工作需要超出本 contract 的范围
29
+ - 实现改动了无关文件
@@ -1,41 +1,41 @@
1
- # 验证矩阵
2
-
3
- 用能证明声明的最窄命令。
4
-
5
- | 变更类型 | 最低验证 | 更强验证 |
6
- |---|---|---|
7
- | 仅文档或治理 | `bash scripts/check-repo.sh` | `bash scripts/ci.sh` |
8
- | TypeScript runtime | `npm run typecheck` | `npm test` |
9
- | CLI 行为 | `npm run typecheck` + 定向 Vitest | `npm test` + CLI smoke |
10
- | DAG 工作流 | 定向 DAG 测试 | `npm test` |
11
- | Production readiness hardening | `bash scripts/check-repo.sh` + 定向 DAG/CLI 测试 | `bash scripts/ci.sh` + docs build + package smoke |
12
- | 脚本或 CI | 运行变更的脚本 | `bash scripts/ci.sh` |
13
- | Package / publish 入口 | `npm run build` + `node bin/loop-agent.js --help` | `npm pack --dry-run` |
14
- | 完整交付 | `bash scripts/ci.sh` | CLI smoke + 相关手工检查 |
15
-
16
- 常用命令:
17
-
18
- ```bash
19
- npm run typecheck
20
- npm test
21
- npm run build
22
- bash scripts/check-repo.sh
23
- bash scripts/ci.sh
24
- node bin/loop-agent.js --help
25
- npm run dev -- --help
26
- npm pack --dry-run
27
- ```
28
-
29
- Windows 上通过 Git Bash 或已配置的兼容 Bash 运行 `scripts/*.sh`。跨平台 CLI 代码对真实文件操作用原生路径;`/` 仅用于稳定 repo 引用、JSON/Markdown 证据引用和 glob 约定。
30
-
31
- 没有相关门禁的新鲜命令输出,不得声明完成。
32
-
33
- Production Readiness v0.1 工作以 `docs/production-readiness.md` 与 `docs/templates/production-readiness-checklist.md` 为验收契约。最终 hardening closeout 需要:
34
-
35
- ```bash
36
- bash scripts/ci.sh
37
- npm run docs:build
38
- npm run build
39
- node bin/loop-agent.js --help
40
- npm pack --dry-run
41
- ```
1
+ # 验证矩阵
2
+
3
+ 用能证明声明的最窄命令。
4
+
5
+ | 变更类型 | 最低验证 | 更强验证 |
6
+ |---|---|---|
7
+ | 仅文档或治理 | `bash scripts/check-repo.sh` | `bash scripts/ci.sh` |
8
+ | TypeScript runtime | `npm run typecheck` | `npm test` |
9
+ | CLI 行为 | `npm run typecheck` + 定向 Vitest | `npm test` + CLI smoke |
10
+ | DAG 工作流 | 定向 DAG 测试 | `npm test` |
11
+ | Production readiness hardening | `bash scripts/check-repo.sh` + 定向 DAG/CLI 测试 | `bash scripts/ci.sh` + docs build + package smoke |
12
+ | 脚本或 CI | 运行变更的脚本 | `bash scripts/ci.sh` |
13
+ | Package / publish 入口 | `npm run build` + `node bin/loop-agent.js --help` | `npm pack --dry-run` |
14
+ | 完整交付 | `bash scripts/ci.sh` | CLI smoke + 相关手工检查 |
15
+
16
+ 常用命令:
17
+
18
+ ```bash
19
+ npm run typecheck
20
+ npm test
21
+ npm run build
22
+ bash scripts/check-repo.sh
23
+ bash scripts/ci.sh
24
+ node bin/loop-agent.js --help
25
+ npm run dev -- --help
26
+ npm pack --dry-run
27
+ ```
28
+
29
+ Windows 上通过 Git Bash 或已配置的兼容 Bash 运行 `scripts/*.sh`。跨平台 CLI 代码对真实文件操作用原生路径;`/` 仅用于稳定 repo 引用、JSON/Markdown 证据引用和 glob 约定。
30
+
31
+ 没有相关门禁的新鲜命令输出,不得声明完成。
32
+
33
+ Production Readiness v0.1 工作以 `docs/production-readiness.md` 与 `docs/templates/production-readiness-checklist.md` 为验收契约。最终 hardening closeout 需要:
34
+
35
+ ```bash
36
+ bash scripts/ci.sh
37
+ npm run docs:build
38
+ npm run build
39
+ node bin/loop-agent.js --help
40
+ npm pack --dry-run
41
+ ```