opencode-swarm-plugin 0.44.0 → 0.44.2

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 (215) hide show
  1. package/bin/swarm.serve.test.ts +6 -4
  2. package/bin/swarm.ts +18 -12
  3. package/dist/compaction-prompt-scoring.js +139 -0
  4. package/dist/eval-capture.js +12811 -0
  5. package/dist/hive.d.ts.map +1 -1
  6. package/dist/hive.js +14834 -0
  7. package/dist/index.d.ts +18 -0
  8. package/dist/index.d.ts.map +1 -1
  9. package/dist/index.js +7743 -62593
  10. package/dist/plugin.js +24052 -78907
  11. package/dist/swarm-orchestrate.d.ts.map +1 -1
  12. package/dist/swarm-prompts.d.ts.map +1 -1
  13. package/dist/swarm-prompts.js +39407 -0
  14. package/dist/swarm-review.d.ts.map +1 -1
  15. package/dist/swarm-validation.d.ts +127 -0
  16. package/dist/swarm-validation.d.ts.map +1 -0
  17. package/dist/validators/index.d.ts +7 -0
  18. package/dist/validators/index.d.ts.map +1 -0
  19. package/dist/validators/schema-validator.d.ts +58 -0
  20. package/dist/validators/schema-validator.d.ts.map +1 -0
  21. package/package.json +17 -5
  22. package/.changeset/swarm-insights-data-layer.md +0 -63
  23. package/.hive/analysis/eval-failure-analysis-2025-12-25.md +0 -331
  24. package/.hive/analysis/session-data-quality-audit.md +0 -320
  25. package/.hive/eval-results.json +0 -483
  26. package/.hive/issues.jsonl +0 -138
  27. package/.hive/memories.jsonl +0 -729
  28. package/.opencode/eval-history.jsonl +0 -327
  29. package/.turbo/turbo-build.log +0 -9
  30. package/CHANGELOG.md +0 -2286
  31. package/SCORER-ANALYSIS.md +0 -598
  32. package/docs/analysis/subagent-coordination-patterns.md +0 -902
  33. package/docs/analysis-socratic-planner-pattern.md +0 -504
  34. package/docs/planning/ADR-001-monorepo-structure.md +0 -171
  35. package/docs/planning/ADR-002-package-extraction.md +0 -393
  36. package/docs/planning/ADR-003-performance-improvements.md +0 -451
  37. package/docs/planning/ADR-004-message-queue-features.md +0 -187
  38. package/docs/planning/ADR-005-devtools-observability.md +0 -202
  39. package/docs/planning/ADR-007-swarm-enhancements-worktree-review.md +0 -168
  40. package/docs/planning/ADR-008-worker-handoff-protocol.md +0 -293
  41. package/docs/planning/ADR-009-oh-my-opencode-patterns.md +0 -353
  42. package/docs/planning/ADR-010-cass-inhousing.md +0 -1215
  43. package/docs/planning/ROADMAP.md +0 -368
  44. package/docs/semantic-memory-cli-syntax.md +0 -123
  45. package/docs/swarm-mail-architecture.md +0 -1147
  46. package/docs/testing/context-recovery-test.md +0 -470
  47. package/evals/ARCHITECTURE.md +0 -1189
  48. package/evals/README.md +0 -768
  49. package/evals/compaction-prompt.eval.ts +0 -149
  50. package/evals/compaction-resumption.eval.ts +0 -289
  51. package/evals/coordinator-behavior.eval.ts +0 -307
  52. package/evals/coordinator-session.eval.ts +0 -154
  53. package/evals/evalite.config.ts.bak +0 -15
  54. package/evals/example.eval.ts +0 -31
  55. package/evals/fixtures/cass-baseline.ts +0 -217
  56. package/evals/fixtures/compaction-cases.ts +0 -350
  57. package/evals/fixtures/compaction-prompt-cases.ts +0 -311
  58. package/evals/fixtures/coordinator-sessions.ts +0 -328
  59. package/evals/fixtures/decomposition-cases.ts +0 -105
  60. package/evals/lib/compaction-loader.test.ts +0 -248
  61. package/evals/lib/compaction-loader.ts +0 -320
  62. package/evals/lib/data-loader.evalite-test.ts +0 -289
  63. package/evals/lib/data-loader.test.ts +0 -345
  64. package/evals/lib/data-loader.ts +0 -281
  65. package/evals/lib/llm.ts +0 -115
  66. package/evals/scorers/compaction-prompt-scorers.ts +0 -145
  67. package/evals/scorers/compaction-scorers.ts +0 -305
  68. package/evals/scorers/coordinator-discipline.evalite-test.ts +0 -539
  69. package/evals/scorers/coordinator-discipline.ts +0 -325
  70. package/evals/scorers/index.test.ts +0 -146
  71. package/evals/scorers/index.ts +0 -328
  72. package/evals/scorers/outcome-scorers.evalite-test.ts +0 -27
  73. package/evals/scorers/outcome-scorers.ts +0 -349
  74. package/evals/swarm-decomposition.eval.ts +0 -121
  75. package/examples/commands/swarm.md +0 -745
  76. package/examples/plugin-wrapper-template.ts +0 -2515
  77. package/examples/skills/hive-workflow/SKILL.md +0 -212
  78. package/examples/skills/skill-creator/SKILL.md +0 -223
  79. package/examples/skills/swarm-coordination/SKILL.md +0 -292
  80. package/global-skills/cli-builder/SKILL.md +0 -344
  81. package/global-skills/cli-builder/references/advanced-patterns.md +0 -244
  82. package/global-skills/learning-systems/SKILL.md +0 -644
  83. package/global-skills/skill-creator/LICENSE.txt +0 -202
  84. package/global-skills/skill-creator/SKILL.md +0 -352
  85. package/global-skills/skill-creator/references/output-patterns.md +0 -82
  86. package/global-skills/skill-creator/references/workflows.md +0 -28
  87. package/global-skills/swarm-coordination/SKILL.md +0 -995
  88. package/global-skills/swarm-coordination/references/coordinator-patterns.md +0 -235
  89. package/global-skills/swarm-coordination/references/strategies.md +0 -138
  90. package/global-skills/system-design/SKILL.md +0 -213
  91. package/global-skills/testing-patterns/SKILL.md +0 -430
  92. package/global-skills/testing-patterns/references/dependency-breaking-catalog.md +0 -586
  93. package/opencode-swarm-plugin-0.30.7.tgz +0 -0
  94. package/opencode-swarm-plugin-0.31.0.tgz +0 -0
  95. package/scripts/cleanup-test-memories.ts +0 -346
  96. package/scripts/init-skill.ts +0 -222
  97. package/scripts/migrate-unknown-sessions.ts +0 -349
  98. package/scripts/validate-skill.ts +0 -204
  99. package/src/agent-mail.ts +0 -1724
  100. package/src/anti-patterns.test.ts +0 -1167
  101. package/src/anti-patterns.ts +0 -448
  102. package/src/compaction-capture.integration.test.ts +0 -257
  103. package/src/compaction-hook.test.ts +0 -838
  104. package/src/compaction-hook.ts +0 -1204
  105. package/src/compaction-observability.integration.test.ts +0 -139
  106. package/src/compaction-observability.test.ts +0 -187
  107. package/src/compaction-observability.ts +0 -324
  108. package/src/compaction-prompt-scorers.test.ts +0 -475
  109. package/src/compaction-prompt-scoring.ts +0 -300
  110. package/src/contributor-tools.test.ts +0 -133
  111. package/src/contributor-tools.ts +0 -201
  112. package/src/dashboard.test.ts +0 -611
  113. package/src/dashboard.ts +0 -462
  114. package/src/error-enrichment.test.ts +0 -403
  115. package/src/error-enrichment.ts +0 -219
  116. package/src/eval-capture.test.ts +0 -1015
  117. package/src/eval-capture.ts +0 -929
  118. package/src/eval-gates.test.ts +0 -306
  119. package/src/eval-gates.ts +0 -218
  120. package/src/eval-history.test.ts +0 -508
  121. package/src/eval-history.ts +0 -214
  122. package/src/eval-learning.test.ts +0 -378
  123. package/src/eval-learning.ts +0 -360
  124. package/src/eval-runner.test.ts +0 -223
  125. package/src/eval-runner.ts +0 -402
  126. package/src/export-tools.test.ts +0 -476
  127. package/src/export-tools.ts +0 -257
  128. package/src/hive.integration.test.ts +0 -2241
  129. package/src/hive.ts +0 -1628
  130. package/src/index.ts +0 -940
  131. package/src/learning.integration.test.ts +0 -1815
  132. package/src/learning.ts +0 -1079
  133. package/src/logger.test.ts +0 -189
  134. package/src/logger.ts +0 -135
  135. package/src/mandate-promotion.test.ts +0 -473
  136. package/src/mandate-promotion.ts +0 -239
  137. package/src/mandate-storage.integration.test.ts +0 -601
  138. package/src/mandate-storage.test.ts +0 -578
  139. package/src/mandate-storage.ts +0 -794
  140. package/src/mandates.ts +0 -540
  141. package/src/memory-tools.test.ts +0 -195
  142. package/src/memory-tools.ts +0 -344
  143. package/src/memory.integration.test.ts +0 -334
  144. package/src/memory.test.ts +0 -158
  145. package/src/memory.ts +0 -527
  146. package/src/model-selection.test.ts +0 -188
  147. package/src/model-selection.ts +0 -68
  148. package/src/observability-tools.test.ts +0 -359
  149. package/src/observability-tools.ts +0 -871
  150. package/src/output-guardrails.test.ts +0 -438
  151. package/src/output-guardrails.ts +0 -381
  152. package/src/pattern-maturity.test.ts +0 -1160
  153. package/src/pattern-maturity.ts +0 -525
  154. package/src/planning-guardrails.test.ts +0 -491
  155. package/src/planning-guardrails.ts +0 -438
  156. package/src/plugin.ts +0 -23
  157. package/src/post-compaction-tracker.test.ts +0 -251
  158. package/src/post-compaction-tracker.ts +0 -237
  159. package/src/query-tools.test.ts +0 -636
  160. package/src/query-tools.ts +0 -324
  161. package/src/rate-limiter.integration.test.ts +0 -466
  162. package/src/rate-limiter.ts +0 -774
  163. package/src/replay-tools.test.ts +0 -496
  164. package/src/replay-tools.ts +0 -240
  165. package/src/repo-crawl.integration.test.ts +0 -441
  166. package/src/repo-crawl.ts +0 -610
  167. package/src/schemas/cell-events.test.ts +0 -347
  168. package/src/schemas/cell-events.ts +0 -807
  169. package/src/schemas/cell.ts +0 -257
  170. package/src/schemas/evaluation.ts +0 -166
  171. package/src/schemas/index.test.ts +0 -199
  172. package/src/schemas/index.ts +0 -286
  173. package/src/schemas/mandate.ts +0 -232
  174. package/src/schemas/swarm-context.ts +0 -115
  175. package/src/schemas/task.ts +0 -161
  176. package/src/schemas/worker-handoff.test.ts +0 -302
  177. package/src/schemas/worker-handoff.ts +0 -131
  178. package/src/sessions/agent-discovery.test.ts +0 -137
  179. package/src/sessions/agent-discovery.ts +0 -112
  180. package/src/sessions/index.ts +0 -15
  181. package/src/skills.integration.test.ts +0 -1192
  182. package/src/skills.test.ts +0 -643
  183. package/src/skills.ts +0 -1549
  184. package/src/storage.integration.test.ts +0 -341
  185. package/src/storage.ts +0 -884
  186. package/src/structured.integration.test.ts +0 -817
  187. package/src/structured.test.ts +0 -1046
  188. package/src/structured.ts +0 -762
  189. package/src/swarm-decompose.test.ts +0 -188
  190. package/src/swarm-decompose.ts +0 -1302
  191. package/src/swarm-deferred.integration.test.ts +0 -157
  192. package/src/swarm-deferred.test.ts +0 -38
  193. package/src/swarm-insights.test.ts +0 -214
  194. package/src/swarm-insights.ts +0 -459
  195. package/src/swarm-mail.integration.test.ts +0 -970
  196. package/src/swarm-mail.ts +0 -739
  197. package/src/swarm-orchestrate.integration.test.ts +0 -282
  198. package/src/swarm-orchestrate.test.ts +0 -548
  199. package/src/swarm-orchestrate.ts +0 -3084
  200. package/src/swarm-prompts.test.ts +0 -1270
  201. package/src/swarm-prompts.ts +0 -2077
  202. package/src/swarm-research.integration.test.ts +0 -701
  203. package/src/swarm-research.test.ts +0 -698
  204. package/src/swarm-research.ts +0 -472
  205. package/src/swarm-review.integration.test.ts +0 -285
  206. package/src/swarm-review.test.ts +0 -879
  207. package/src/swarm-review.ts +0 -709
  208. package/src/swarm-strategies.ts +0 -407
  209. package/src/swarm-worktree.test.ts +0 -501
  210. package/src/swarm-worktree.ts +0 -575
  211. package/src/swarm.integration.test.ts +0 -2377
  212. package/src/swarm.ts +0 -38
  213. package/src/tool-adapter.integration.test.ts +0 -1221
  214. package/src/tool-availability.ts +0 -461
  215. package/tsconfig.json +0 -28
@@ -1,289 +0,0 @@
1
- /**
2
- * Data Loader Tests
3
- *
4
- * Tests the PGlite-backed eval data loader functions.
5
- * Uses a real in-memory PGlite database for accurate testing.
6
- */
7
- import { describe, it, expect, beforeAll, afterAll } from "bun:test";
8
- import {
9
- loadEvalCases,
10
- hasRealEvalData,
11
- getEvalDataSummary,
12
- } from "./data-loader.js";
13
- import {
14
- appendEvent,
15
- getDatabase,
16
- closeDatabase,
17
- type DecompositionGeneratedEvent,
18
- type SubtaskOutcomeEvent,
19
- } from "swarm-mail";
20
- import * as fs from "node:fs";
21
- import * as path from "node:path";
22
- import * as os from "node:os";
23
-
24
- const TEST_PROJECT_KEY = "test-project-eval-loader";
25
-
26
- // Create a unique temp directory for this test run
27
- let testDir: string;
28
-
29
- describe("Data Loader", () => {
30
- beforeAll(async () => {
31
- // Create temp directory for test database
32
- testDir = fs.mkdtempSync(path.join(os.tmpdir(), "eval-loader-test-"));
33
-
34
- // Initialize database by getting it (lazy init)
35
- await getDatabase(testDir);
36
- });
37
-
38
- afterAll(async () => {
39
- await closeDatabase(testDir);
40
- // Clean up temp directory
41
- fs.rmSync(testDir, { recursive: true, force: true });
42
- });
43
-
44
- describe("loadEvalCases", () => {
45
- it("transforms eval records to EvalCase format", async () => {
46
- // Insert a decomposition event
47
- const decompositionEvent: DecompositionGeneratedEvent = {
48
- type: "decomposition_generated",
49
- timestamp: Date.now(),
50
- project_key: TEST_PROJECT_KEY,
51
- epic_id: "epic-load-1",
52
- task: "Add authentication",
53
- context: "Next.js app",
54
- strategy: "feature-based",
55
- epic_title: "Auth Epic",
56
- subtasks: [
57
- { title: "OAuth setup", files: ["src/auth/oauth.ts"], priority: 1 },
58
- {
59
- title: "Session management",
60
- files: ["src/auth/session.ts"],
61
- priority: 2,
62
- },
63
- ],
64
- };
65
- await appendEvent(decompositionEvent, testDir);
66
-
67
- // Insert outcome events for both subtasks
68
- const outcome1: SubtaskOutcomeEvent = {
69
- type: "subtask_outcome",
70
- timestamp: Date.now(),
71
- project_key: TEST_PROJECT_KEY,
72
- epic_id: "epic-load-1",
73
- bead_id: "epic-load-1.1",
74
- planned_files: ["src/auth/oauth.ts"],
75
- actual_files: ["src/auth/oauth.ts"],
76
- duration_ms: 5000,
77
- error_count: 0,
78
- retry_count: 0,
79
- success: true,
80
- };
81
- await appendEvent(outcome1, testDir);
82
-
83
- const outcome2: SubtaskOutcomeEvent = {
84
- type: "subtask_outcome",
85
- timestamp: Date.now(),
86
- project_key: TEST_PROJECT_KEY,
87
- epic_id: "epic-load-1",
88
- bead_id: "epic-load-1.2",
89
- planned_files: ["src/auth/session.ts"],
90
- actual_files: ["src/auth/session.ts"],
91
- duration_ms: 3000,
92
- error_count: 0,
93
- retry_count: 0,
94
- success: true,
95
- };
96
- await appendEvent(outcome2, testDir);
97
-
98
- const cases = await loadEvalCases(TEST_PROJECT_KEY, {
99
- projectPath: testDir,
100
- });
101
-
102
- expect(cases.length).toBeGreaterThanOrEqual(1);
103
- const authCase = cases.find((c) => c.input.task === "Add authentication");
104
- expect(authCase).toBeDefined();
105
- expect(authCase!.input.context).toBe("Next.js app");
106
- expect(authCase!.expected.minSubtasks).toBe(2);
107
- expect(authCase!.expected.maxSubtasks).toBe(2);
108
- expect(authCase!.expected.requiredFiles).toContain("src/auth/oauth.ts");
109
- expect(authCase!.expected.requiredFiles).toContain("src/auth/session.ts");
110
- expect(authCase!.actual).toBeDefined();
111
- });
112
-
113
- it("filters by success when successOnly is true", async () => {
114
- // Insert a successful decomposition
115
- const successEvent: DecompositionGeneratedEvent = {
116
- type: "decomposition_generated",
117
- timestamp: Date.now(),
118
- project_key: TEST_PROJECT_KEY,
119
- epic_id: "epic-success-filter",
120
- task: "Success task for filter",
121
- strategy: "feature-based",
122
- epic_title: "Success Epic",
123
- subtasks: [{ title: "Sub", files: ["src/success.ts"], priority: 1 }],
124
- };
125
- await appendEvent(successEvent, testDir);
126
-
127
- // Mark it successful
128
- const successOutcome: SubtaskOutcomeEvent = {
129
- type: "subtask_outcome",
130
- timestamp: Date.now(),
131
- project_key: TEST_PROJECT_KEY,
132
- epic_id: "epic-success-filter",
133
- bead_id: "epic-success-filter.1",
134
- planned_files: ["src/success.ts"],
135
- actual_files: ["src/success.ts"],
136
- duration_ms: 1000,
137
- error_count: 0,
138
- retry_count: 0,
139
- success: true,
140
- };
141
- await appendEvent(successOutcome, testDir);
142
-
143
- // Insert a failed decomposition
144
- const failEvent: DecompositionGeneratedEvent = {
145
- type: "decomposition_generated",
146
- timestamp: Date.now(),
147
- project_key: TEST_PROJECT_KEY,
148
- epic_id: "epic-fail-filter",
149
- task: "Failed task for filter",
150
- strategy: "feature-based",
151
- epic_title: "Failed Epic",
152
- subtasks: [{ title: "Sub", files: ["src/fail.ts"], priority: 1 }],
153
- };
154
- await appendEvent(failEvent, testDir);
155
-
156
- // Mark it failed
157
- const failOutcome: SubtaskOutcomeEvent = {
158
- type: "subtask_outcome",
159
- timestamp: Date.now(),
160
- project_key: TEST_PROJECT_KEY,
161
- epic_id: "epic-fail-filter",
162
- bead_id: "epic-fail-filter.1",
163
- planned_files: ["src/fail.ts"],
164
- actual_files: [],
165
- duration_ms: 500,
166
- error_count: 3,
167
- retry_count: 2,
168
- success: false,
169
- };
170
- await appendEvent(failOutcome, testDir);
171
-
172
- const successCases = await loadEvalCases(TEST_PROJECT_KEY, {
173
- successOnly: true,
174
- projectPath: testDir,
175
- });
176
-
177
- // Should only include successful cases
178
- const failedCase = successCases.find(
179
- (c) => c.input.task === "Failed task for filter",
180
- );
181
- expect(failedCase).toBeUndefined();
182
- });
183
-
184
- it("passes strategy filter to getEvalRecords", async () => {
185
- // Insert file-based decomposition
186
- const fileBasedEvent: DecompositionGeneratedEvent = {
187
- type: "decomposition_generated",
188
- timestamp: Date.now(),
189
- project_key: TEST_PROJECT_KEY,
190
- epic_id: "epic-file-based",
191
- task: "File-based task",
192
- strategy: "file-based",
193
- epic_title: "File Epic",
194
- subtasks: [{ title: "Sub", files: ["src/file.ts"], priority: 1 }],
195
- };
196
- await appendEvent(fileBasedEvent, testDir);
197
-
198
- const fileBasedCases = await loadEvalCases(TEST_PROJECT_KEY, {
199
- strategy: "file-based",
200
- projectPath: testDir,
201
- });
202
-
203
- // All returned cases should be file-based
204
- for (const c of fileBasedCases) {
205
- expect(c.actual?.strategy).toBe("file-based");
206
- }
207
- });
208
-
209
- it("passes limit to getEvalRecords", async () => {
210
- const cases = await loadEvalCases(TEST_PROJECT_KEY, {
211
- limit: 2,
212
- projectPath: testDir,
213
- });
214
-
215
- expect(cases.length).toBeLessThanOrEqual(2);
216
- });
217
-
218
- it("handles records with no context", async () => {
219
- const noContextEvent: DecompositionGeneratedEvent = {
220
- type: "decomposition_generated",
221
- timestamp: Date.now(),
222
- project_key: TEST_PROJECT_KEY,
223
- epic_id: "epic-no-context",
224
- task: "Task without context",
225
- // context is undefined
226
- strategy: "feature-based",
227
- epic_title: "No Context Epic",
228
- subtasks: [{ title: "Sub", files: [], priority: 1 }],
229
- };
230
- await appendEvent(noContextEvent, testDir);
231
-
232
- const cases = await loadEvalCases(TEST_PROJECT_KEY, {
233
- projectPath: testDir,
234
- });
235
- const noContextCase = cases.find(
236
- (c) => c.input.task === "Task without context",
237
- );
238
-
239
- expect(noContextCase).toBeDefined();
240
- expect(noContextCase!.input.context).toBeUndefined();
241
- });
242
- });
243
-
244
- describe("hasRealEvalData", () => {
245
- it("returns true when enough records exist", async () => {
246
- // We've inserted several records above, should have enough
247
- const hasData = await hasRealEvalData(TEST_PROJECT_KEY, 1, testDir);
248
- expect(hasData).toBe(true);
249
- });
250
-
251
- it("returns false when not enough records exist", async () => {
252
- // Use a project key with no data
253
- const hasData = await hasRealEvalData("nonexistent-project", 5, testDir);
254
- expect(hasData).toBe(false);
255
- });
256
-
257
- it("uses custom minRecords threshold", async () => {
258
- // Should have at least 1 record
259
- const hasData = await hasRealEvalData(TEST_PROJECT_KEY, 1, testDir);
260
- expect(hasData).toBe(true);
261
-
262
- // Should not have 1000 records
263
- const hasLotsOfData = await hasRealEvalData(
264
- TEST_PROJECT_KEY,
265
- 1000,
266
- testDir,
267
- );
268
- expect(hasLotsOfData).toBe(false);
269
- });
270
- });
271
-
272
- describe("getEvalDataSummary", () => {
273
- it("returns formatted summary with hasEnoughData flag", async () => {
274
- const summary = await getEvalDataSummary(TEST_PROJECT_KEY, testDir);
275
-
276
- expect(summary.totalRecords).toBeGreaterThanOrEqual(1);
277
- expect(typeof summary.successRate).toBe("number");
278
- expect(typeof summary.byStrategy).toBe("object");
279
- expect(typeof summary.hasEnoughData).toBe("boolean");
280
- });
281
-
282
- it("sets hasEnoughData based on record count", async () => {
283
- // Empty project should not have enough data
284
- const emptySummary = await getEvalDataSummary("empty-project", testDir);
285
- expect(emptySummary.hasEnoughData).toBe(false);
286
- expect(emptySummary.totalRecords).toBe(0);
287
- });
288
- });
289
- });
@@ -1,345 +0,0 @@
1
- /**
2
- * Tests for data-loader quality filters
3
- *
4
- * TDD approach: RED → GREEN → REFACTOR
5
- */
6
- import { afterEach, beforeEach, describe, expect, test } from "bun:test";
7
- import * as fs from "node:fs";
8
- import * as os from "node:os";
9
- import * as path from "node:path";
10
- import type { CoordinatorEvent } from "../../src/eval-capture.js";
11
- import { loadCapturedSessions } from "./data-loader.js";
12
-
13
- // Test helper: create a temp session directory
14
- let tempSessionDir: string;
15
-
16
- beforeEach(() => {
17
- tempSessionDir = fs.mkdtempSync(path.join(os.tmpdir(), "test-sessions-"));
18
- });
19
-
20
- afterEach(() => {
21
- if (fs.existsSync(tempSessionDir)) {
22
- fs.rmSync(tempSessionDir, { recursive: true });
23
- }
24
- });
25
-
26
- /**
27
- * Helper: create a session JSONL file with events
28
- */
29
- function createSessionFile(
30
- sessionId: string,
31
- events: CoordinatorEvent[],
32
- ): void {
33
- const filePath = path.join(tempSessionDir, `${sessionId}.jsonl`);
34
- const lines = events.map((e) => JSON.stringify(e)).join("\n") + "\n";
35
- fs.writeFileSync(filePath, lines, "utf-8");
36
- }
37
-
38
- /**
39
- * Helper: create minimal events
40
- */
41
- function createEvent(
42
- sessionId: string,
43
- epicId: string,
44
- type: "DECISION" | "VIOLATION" | "OUTCOME",
45
- subtype: string,
46
- ): CoordinatorEvent {
47
- const base = {
48
- session_id: sessionId,
49
- epic_id: epicId,
50
- timestamp: new Date().toISOString(),
51
- payload: {},
52
- };
53
-
54
- if (type === "DECISION") {
55
- return {
56
- ...base,
57
- event_type: "DECISION" as const,
58
- decision_type: subtype as any,
59
- };
60
- } else if (type === "VIOLATION") {
61
- return {
62
- ...base,
63
- event_type: "VIOLATION" as const,
64
- violation_type: subtype as any,
65
- };
66
- } else {
67
- return {
68
- ...base,
69
- event_type: "OUTCOME" as const,
70
- outcome_type: subtype as any,
71
- };
72
- }
73
- }
74
-
75
- describe("loadCapturedSessions - quality filters", () => {
76
- test("filters out sessions with fewer than minEvents (default: 3)", async () => {
77
- // Create sessions with different event counts
78
- createSessionFile("session-2-events", [
79
- createEvent("session-2-events", "epic-1", "DECISION", "worker_spawned"),
80
- createEvent("session-2-events", "epic-1", "OUTCOME", "subtask_success"),
81
- ]);
82
-
83
- createSessionFile("session-3-events", [
84
- createEvent("session-3-events", "epic-2", "DECISION", "worker_spawned"),
85
- createEvent("session-3-events", "epic-2", "DECISION", "review_completed"),
86
- createEvent("session-3-events", "epic-2", "OUTCOME", "subtask_success"),
87
- ]);
88
-
89
- createSessionFile("session-5-events", [
90
- createEvent("session-5-events", "epic-3", "DECISION", "worker_spawned"),
91
- createEvent("session-5-events", "epic-3", "DECISION", "review_completed"),
92
- createEvent("session-5-events", "epic-3", "OUTCOME", "subtask_success"),
93
- createEvent("session-5-events", "epic-3", "OUTCOME", "subtask_success"),
94
- createEvent("session-5-events", "epic-3", "OUTCOME", "epic_complete"),
95
- ]);
96
-
97
- const sessions = await loadCapturedSessions({
98
- minEvents: 3,
99
- sessionDir: tempSessionDir,
100
- });
101
-
102
- // Should only get sessions with >= 3 events
103
- expect(sessions.length).toBe(2);
104
- expect(
105
- sessions.some((s) => s.session.session_id === "session-3-events"),
106
- ).toBe(true);
107
- expect(
108
- sessions.some((s) => s.session.session_id === "session-5-events"),
109
- ).toBe(true);
110
- expect(
111
- sessions.some((s) => s.session.session_id === "session-2-events"),
112
- ).toBe(false);
113
- });
114
-
115
- test("filters out sessions without worker_spawned event when requireWorkerSpawn=true", async () => {
116
- // Session WITH worker_spawned
117
- createSessionFile("session-with-spawn", [
118
- createEvent("session-with-spawn", "epic-1", "DECISION", "worker_spawned"),
119
- createEvent(
120
- "session-with-spawn",
121
- "epic-1",
122
- "DECISION",
123
- "review_completed",
124
- ),
125
- createEvent("session-with-spawn", "epic-1", "OUTCOME", "subtask_success"),
126
- ]);
127
-
128
- // Session WITHOUT worker_spawned
129
- createSessionFile("session-no-spawn", [
130
- createEvent(
131
- "session-no-spawn",
132
- "epic-2",
133
- "DECISION",
134
- "strategy_selected",
135
- ),
136
- createEvent(
137
- "session-no-spawn",
138
- "epic-2",
139
- "DECISION",
140
- "decomposition_complete",
141
- ),
142
- createEvent("session-no-spawn", "epic-2", "OUTCOME", "epic_complete"),
143
- ]);
144
-
145
- const sessions = await loadCapturedSessions({
146
- requireWorkerSpawn: true,
147
- sessionDir: tempSessionDir,
148
- });
149
-
150
- expect(sessions.length).toBe(1);
151
- expect(sessions[0]?.session.session_id).toBe("session-with-spawn");
152
- });
153
-
154
- test("filters out sessions without review_completed event when requireReview=true", async () => {
155
- // Session WITH review
156
- createSessionFile("session-with-review", [
157
- createEvent(
158
- "session-with-review",
159
- "epic-1",
160
- "DECISION",
161
- "worker_spawned",
162
- ),
163
- createEvent(
164
- "session-with-review",
165
- "epic-1",
166
- "DECISION",
167
- "review_completed",
168
- ),
169
- createEvent("session-with-review", "epic-1", "OUTCOME", "subtask_success"),
170
- ]);
171
-
172
- // Session WITHOUT review
173
- createSessionFile("session-no-review", [
174
- createEvent("session-no-review", "epic-2", "DECISION", "worker_spawned"),
175
- createEvent("session-no-review", "epic-2", "OUTCOME", "subtask_success"),
176
- createEvent("session-no-review", "epic-2", "OUTCOME", "epic_complete"),
177
- ]);
178
-
179
- const sessions = await loadCapturedSessions({
180
- requireReview: true,
181
- sessionDir: tempSessionDir,
182
- });
183
-
184
- expect(sessions.length).toBe(1);
185
- expect(sessions[0]?.session.session_id).toBe("session-with-review");
186
- });
187
-
188
- test("allows disabling filters individually", async () => {
189
- // Session with only 2 events, no worker_spawned, no review
190
- createSessionFile("session-low-quality", [
191
- createEvent(
192
- "session-low-quality",
193
- "epic-1",
194
- "DECISION",
195
- "strategy_selected",
196
- ),
197
- createEvent("session-low-quality", "epic-1", "OUTCOME", "epic_complete"),
198
- ]);
199
-
200
- // Disable all filters
201
- const sessions = await loadCapturedSessions({
202
- minEvents: 0,
203
- requireWorkerSpawn: false,
204
- requireReview: false,
205
- sessionDir: tempSessionDir,
206
- });
207
-
208
- expect(sessions.length).toBe(1);
209
- expect(sessions[0]?.session.session_id).toBe("session-low-quality");
210
- });
211
-
212
- test("applies limit AFTER filtering", async () => {
213
- // Create 5 high-quality sessions
214
- for (let i = 1; i <= 5; i++) {
215
- createSessionFile(`session-${i}`, [
216
- createEvent(`session-${i}`, `epic-${i}`, "DECISION", "worker_spawned"),
217
- createEvent(
218
- `session-${i}`,
219
- `epic-${i}`,
220
- "DECISION",
221
- "review_completed",
222
- ),
223
- createEvent(`session-${i}`, `epic-${i}`, "OUTCOME", "subtask_success"),
224
- ]);
225
- }
226
-
227
- // Create 3 low-quality sessions (will be filtered out)
228
- for (let i = 6; i <= 8; i++) {
229
- createSessionFile(`session-${i}`, [
230
- createEvent(`session-${i}`, `epic-${i}`, "DECISION", "strategy_selected"),
231
- ]);
232
- }
233
-
234
- // Filter first (remove 3 low-quality), then limit to 2
235
- const sessions = await loadCapturedSessions({
236
- minEvents: 3,
237
- requireWorkerSpawn: true,
238
- requireReview: true,
239
- limit: 2,
240
- sessionDir: tempSessionDir,
241
- });
242
-
243
- // Should get 2 sessions from the 5 high-quality ones
244
- expect(sessions.length).toBe(2);
245
- expect(sessions.every((s) => s.session.events.length >= 3)).toBe(true);
246
- });
247
-
248
- test("combines all filters correctly", async () => {
249
- // High-quality session (passes all filters)
250
- createSessionFile("session-high-quality", [
251
- createEvent(
252
- "session-high-quality",
253
- "epic-1",
254
- "DECISION",
255
- "worker_spawned",
256
- ),
257
- createEvent(
258
- "session-high-quality",
259
- "epic-1",
260
- "DECISION",
261
- "review_completed",
262
- ),
263
- createEvent("session-high-quality", "epic-1", "OUTCOME", "subtask_success"),
264
- createEvent("session-high-quality", "epic-1", "OUTCOME", "epic_complete"),
265
- ]);
266
-
267
- // Missing worker_spawned
268
- createSessionFile("session-no-spawn", [
269
- createEvent(
270
- "session-no-spawn",
271
- "epic-2",
272
- "DECISION",
273
- "review_completed",
274
- ),
275
- createEvent("session-no-spawn", "epic-2", "OUTCOME", "subtask_success"),
276
- createEvent("session-no-spawn", "epic-2", "OUTCOME", "epic_complete"),
277
- ]);
278
-
279
- // Missing review_completed
280
- createSessionFile("session-no-review", [
281
- createEvent("session-no-review", "epic-3", "DECISION", "worker_spawned"),
282
- createEvent("session-no-review", "epic-3", "OUTCOME", "subtask_success"),
283
- createEvent("session-no-review", "epic-3", "OUTCOME", "epic_complete"),
284
- ]);
285
-
286
- // Too few events
287
- createSessionFile("session-too-few", [
288
- createEvent("session-too-few", "epic-4", "DECISION", "worker_spawned"),
289
- createEvent("session-too-few", "epic-4", "DECISION", "review_completed"),
290
- ]);
291
-
292
- const sessions = await loadCapturedSessions({
293
- minEvents: 3,
294
- requireWorkerSpawn: true,
295
- requireReview: true,
296
- sessionDir: tempSessionDir,
297
- });
298
-
299
- // Only high-quality session should pass
300
- expect(sessions.length).toBe(1);
301
- expect(sessions[0]?.session.session_id).toBe("session-high-quality");
302
- });
303
-
304
- test("defaults are: minEvents=3, requireWorkerSpawn=true, requireReview=true", async () => {
305
- // Create one session that meets defaults
306
- createSessionFile("session-meets-defaults", [
307
- createEvent(
308
- "session-meets-defaults",
309
- "epic-1",
310
- "DECISION",
311
- "worker_spawned",
312
- ),
313
- createEvent(
314
- "session-meets-defaults",
315
- "epic-1",
316
- "DECISION",
317
- "review_completed",
318
- ),
319
- createEvent(
320
- "session-meets-defaults",
321
- "epic-1",
322
- "OUTCOME",
323
- "subtask_success",
324
- ),
325
- ]);
326
-
327
- // Create one that doesn't
328
- createSessionFile("session-fails-defaults", [
329
- createEvent(
330
- "session-fails-defaults",
331
- "epic-2",
332
- "DECISION",
333
- "strategy_selected",
334
- ),
335
- ]);
336
-
337
- // Call with NO options except sessionDir - should use defaults
338
- const sessions = await loadCapturedSessions({
339
- sessionDir: tempSessionDir,
340
- });
341
-
342
- expect(sessions.length).toBe(1);
343
- expect(sessions[0]?.session.session_id).toBe("session-meets-defaults");
344
- });
345
- });