@lesliechan721/agent-skill-eval 0.1.0-beta.1 → 0.1.0-beta.10

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 (101) hide show
  1. package/dist/bin.js +167 -39
  2. package/dist/bin.js.map +1 -1
  3. package/dist/codex-eval-app-server.d.ts +199 -0
  4. package/dist/codex-eval-app-server.js +1364 -0
  5. package/dist/codex-eval-app-server.js.map +1 -0
  6. package/dist/codex-eval-cases.d.ts +43 -0
  7. package/dist/codex-eval-cases.js +327 -0
  8. package/dist/codex-eval-cases.js.map +1 -0
  9. package/dist/codex-eval-isolation.d.ts +79 -0
  10. package/dist/codex-eval-isolation.js +420 -0
  11. package/dist/codex-eval-isolation.js.map +1 -0
  12. package/dist/codex-eval-process-tree.d.ts +38 -0
  13. package/dist/codex-eval-process-tree.js +803 -0
  14. package/dist/codex-eval-process-tree.js.map +1 -0
  15. package/dist/codex-eval-runtime-config.d.ts +43 -0
  16. package/dist/codex-eval-runtime-config.js +607 -0
  17. package/dist/codex-eval-runtime-config.js.map +1 -0
  18. package/dist/codex-trigger-eval.d.ts +82 -0
  19. package/dist/codex-trigger-eval.js +790 -0
  20. package/dist/codex-trigger-eval.js.map +1 -0
  21. package/dist/codex-workflow.d.ts +191 -0
  22. package/dist/codex-workflow.js +1832 -0
  23. package/dist/codex-workflow.js.map +1 -0
  24. package/dist/credentials.d.ts +5 -0
  25. package/dist/credentials.js +30 -0
  26. package/dist/credentials.js.map +1 -0
  27. package/dist/custom-process.d.ts +60 -0
  28. package/dist/custom-process.js +423 -0
  29. package/dist/custom-process.js.map +1 -0
  30. package/dist/eval-audit.d.ts +38 -0
  31. package/dist/eval-audit.js +71 -0
  32. package/dist/eval-audit.js.map +1 -0
  33. package/dist/eval-file.d.ts +9 -0
  34. package/dist/eval-file.js +15 -0
  35. package/dist/eval-file.js.map +1 -0
  36. package/dist/eval-process-management.d.ts +21 -0
  37. package/dist/eval-process-management.js +84 -0
  38. package/dist/eval-process-management.js.map +1 -0
  39. package/dist/eval-result.d.ts +79 -0
  40. package/dist/eval-result.js +144 -0
  41. package/dist/eval-result.js.map +1 -0
  42. package/dist/eval-run.d.ts +27 -0
  43. package/dist/eval-run.js +564 -0
  44. package/dist/eval-run.js.map +1 -0
  45. package/dist/eval-spec.d.ts +57 -0
  46. package/dist/eval-spec.js +120 -0
  47. package/dist/eval-spec.js.map +1 -0
  48. package/dist/eval-suite-binding.d.ts +4 -0
  49. package/dist/eval-suite-binding.js +19 -0
  50. package/dist/eval-suite-binding.js.map +1 -0
  51. package/dist/index.d.ts +36 -9
  52. package/dist/index.js +19 -6
  53. package/dist/index.js.map +1 -1
  54. package/dist/json-schema.js +75 -5
  55. package/dist/json-schema.js.map +1 -1
  56. package/dist/output-path.js +2 -2
  57. package/dist/process-tree.d.ts +19 -1
  58. package/dist/process-tree.js +391 -77
  59. package/dist/process-tree.js.map +1 -1
  60. package/dist/process.d.ts +0 -2
  61. package/dist/process.js +76 -25
  62. package/dist/process.js.map +1 -1
  63. package/dist/provider-doctor.d.ts +45 -0
  64. package/dist/provider-doctor.js +112 -0
  65. package/dist/provider-doctor.js.map +1 -0
  66. package/dist/providers.d.ts +83 -0
  67. package/dist/providers.js +632 -0
  68. package/dist/providers.js.map +1 -0
  69. package/dist/run-paths.d.ts +8 -0
  70. package/dist/run-paths.js +71 -0
  71. package/dist/run-paths.js.map +1 -0
  72. package/dist/runtime-skill.d.ts +6 -0
  73. package/dist/runtime-skill.js +87 -3
  74. package/dist/runtime-skill.js.map +1 -1
  75. package/dist/schemas/digests.json +3 -3
  76. package/dist/schemas/eval-audit.schema.json +64 -0
  77. package/dist/schemas/eval-result.schema.json +232 -0
  78. package/dist/schemas/eval-spec.schema.json +109 -0
  79. package/dist/schemas.d.ts +1 -3
  80. package/dist/schemas.js +0 -17
  81. package/dist/schemas.js.map +1 -1
  82. package/dist/selected-cases.d.ts +30 -0
  83. package/dist/selected-cases.js +388 -0
  84. package/dist/selected-cases.js.map +1 -0
  85. package/dist/types.d.ts +2 -34
  86. package/package.json +8 -3
  87. package/dist/bundle.d.ts +0 -93
  88. package/dist/bundle.js +0 -216
  89. package/dist/bundle.js.map +0 -1
  90. package/dist/capabilities.d.ts +0 -2
  91. package/dist/capabilities.js +0 -38
  92. package/dist/capabilities.js.map +0 -1
  93. package/dist/plan.d.ts +0 -79
  94. package/dist/plan.js +0 -285
  95. package/dist/plan.js.map +0 -1
  96. package/dist/run.d.ts +0 -2
  97. package/dist/run.js +0 -592
  98. package/dist/run.js.map +0 -1
  99. package/dist/schemas/aw-skill-eval-execution-plan.schema.json +0 -220
  100. package/dist/schemas/eval-run-bundle.schema.json +0 -408
  101. package/dist/schemas/eval-runtime-capabilities.schema.json +0 -85
@@ -0,0 +1,82 @@
1
+ import { type CodexEvalProcessManagementEvidence } from './codex-eval-app-server.js';
2
+ import { type CodexEvalRuntimeConfig } from './codex-eval-runtime-config.js';
3
+ import { type AwSkillEvalTriggerCase, type AwSkillEvalTriggerSuiteCaseContract } from './codex-eval-cases.js';
4
+ interface TriggerObservation {
5
+ loaded: boolean;
6
+ spawned: boolean;
7
+ loadEvidence: string | null;
8
+ spawnEvidence: string | null;
9
+ }
10
+ export interface TriggerRunEvidence {
11
+ loaded: boolean;
12
+ first_loaded_skill: string | null;
13
+ all_loaded_skills: string[];
14
+ load_evidence_by_skill: Record<string, string>;
15
+ spawned: boolean;
16
+ load_evidence: string | null;
17
+ spawn_evidence: string | null;
18
+ spawn_calls: Array<Record<string, unknown>>;
19
+ termination: string;
20
+ exit_code: number | null;
21
+ signal: NodeJS.Signals | null;
22
+ execution_error: boolean;
23
+ fixture_unchanged: boolean;
24
+ request_user_input_called: boolean;
25
+ waiting_on_user_input: boolean;
26
+ final_response_emitted: boolean;
27
+ stderr?: string;
28
+ final_message: string | null;
29
+ duration_ms: number;
30
+ usage: unknown;
31
+ event_trace: unknown[];
32
+ diagnostics: string[];
33
+ }
34
+ interface TriggerCaseSummary {
35
+ id: string;
36
+ split: AwSkillEvalTriggerCase['split'];
37
+ query: string;
38
+ expected: {
39
+ load_skill?: boolean;
40
+ loaded_skills?: string[];
41
+ first_loaded_skill?: string;
42
+ spawn_agent?: boolean;
43
+ };
44
+ observed: {
45
+ load_rate: number;
46
+ skill_load_rates: Record<string, number>;
47
+ route_match_rate?: number;
48
+ first_route_match_rate?: number;
49
+ route_counts?: Record<string, number>;
50
+ multiple_skill_loads: number;
51
+ spawn_rate?: number;
52
+ execution_errors: number;
53
+ };
54
+ complete: boolean;
55
+ passed: boolean;
56
+ runs: TriggerRunEvidence[];
57
+ }
58
+ export declare function inspectTriggerAppServerFrame(frame: Record<string, unknown>, skillPath: string): Partial<TriggerObservation> & {
59
+ finalMessage?: string;
60
+ usage?: unknown;
61
+ turnStatus?: string;
62
+ spawnCall?: Record<string, unknown>;
63
+ requestUserInput?: boolean;
64
+ };
65
+ export declare function inspectTriggerAppServerConfigWarning(frame: Record<string, unknown>): string | null;
66
+ export declare function runCodexTriggerProviderCase(options: {
67
+ suite: AwSkillEvalTriggerSuiteCaseContract;
68
+ evalCase: AwSkillEvalTriggerCase;
69
+ subjectRoot: string;
70
+ fixtureRoot: string;
71
+ codexBin: string;
72
+ model?: string;
73
+ codexRuntime: CodexEvalRuntimeConfig;
74
+ sourceEnv?: NodeJS.ProcessEnv;
75
+ processManagement?: CodexEvalProcessManagementEvidence;
76
+ workRoot?: string;
77
+ manageTimeout?: boolean;
78
+ signal?: AbortSignal;
79
+ timeoutMs?: number;
80
+ }): Promise<TriggerRunEvidence>;
81
+ export declare function summarizeTriggerCase(suite: AwSkillEvalTriggerSuiteCaseContract, evalCase: AwSkillEvalTriggerCase, runs: TriggerRunEvidence[]): TriggerCaseSummary;
82
+ export {};