@bradtaylorsf/alpha-loop 1.9.1 → 1.10.0

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 (53) hide show
  1. package/README.md +10 -1
  2. package/dist/cli.js +15 -3
  3. package/dist/cli.js.map +1 -1
  4. package/dist/commands/eval.d.ts +13 -0
  5. package/dist/commands/eval.js +220 -3
  6. package/dist/commands/eval.js.map +1 -1
  7. package/dist/commands/init.js +32 -0
  8. package/dist/commands/init.js.map +1 -1
  9. package/dist/commands/review.js +36 -1
  10. package/dist/commands/review.js.map +1 -1
  11. package/dist/lib/config.d.ts +8 -0
  12. package/dist/lib/config.js +18 -0
  13. package/dist/lib/config.js.map +1 -1
  14. package/dist/lib/eval-export.d.ts +26 -0
  15. package/dist/lib/eval-export.js +200 -0
  16. package/dist/lib/eval-export.js.map +1 -0
  17. package/dist/lib/eval-runner.d.ts +11 -0
  18. package/dist/lib/eval-runner.js +122 -10
  19. package/dist/lib/eval-runner.js.map +1 -1
  20. package/dist/lib/eval.d.ts +11 -0
  21. package/dist/lib/eval.js +34 -4
  22. package/dist/lib/eval.js.map +1 -1
  23. package/dist/lib/github.js +20 -12
  24. package/dist/lib/github.js.map +1 -1
  25. package/dist/lib/pipeline.d.ts +4 -0
  26. package/dist/lib/pipeline.js +54 -8
  27. package/dist/lib/pipeline.js.map +1 -1
  28. package/dist/lib/rate-limit.js +12 -5
  29. package/dist/lib/rate-limit.js.map +1 -1
  30. package/dist/lib/score.js +2 -1
  31. package/dist/lib/score.js.map +1 -1
  32. package/dist/lib/templates.d.ts +11 -0
  33. package/dist/lib/templates.js +27 -0
  34. package/dist/lib/templates.js.map +1 -1
  35. package/dist/lib/verify.d.ts +16 -0
  36. package/dist/lib/verify.js +49 -1
  37. package/dist/lib/verify.js.map +1 -1
  38. package/package.json +1 -1
  39. package/templates/evals/cases/step/review/006-missing-di-injection/checks.yaml +10 -0
  40. package/templates/evals/cases/step/review/006-missing-di-injection/input.md +50 -0
  41. package/templates/evals/cases/step/review/006-missing-di-injection/metadata.yaml +10 -0
  42. package/templates/evals/cases/step/review/007-silent-none-guard/checks.yaml +12 -0
  43. package/templates/evals/cases/step/review/007-silent-none-guard/input.md +53 -0
  44. package/templates/evals/cases/step/review/007-silent-none-guard/metadata.yaml +10 -0
  45. package/templates/evals/cases/step/review/008-route-shadowing/checks.yaml +11 -0
  46. package/templates/evals/cases/step/review/008-route-shadowing/input.md +29 -0
  47. package/templates/evals/cases/step/review/008-route-shadowing/metadata.yaml +10 -0
  48. package/templates/evals/cases/step/review/009-unthreaded-dependency/checks.yaml +11 -0
  49. package/templates/evals/cases/step/review/009-unthreaded-dependency/input.md +57 -0
  50. package/templates/evals/cases/step/review/009-unthreaded-dependency/metadata.yaml +10 -0
  51. package/templates/evals/cases/step/review/010-fabricated-metrics/checks.yaml +12 -0
  52. package/templates/evals/cases/step/review/010-fabricated-metrics/input.md +52 -0
  53. package/templates/evals/cases/step/review/010-fabricated-metrics/metadata.yaml +10 -0
@@ -0,0 +1,52 @@
1
+ ## Issue: Add cost tracking to pipeline runs
2
+
3
+ ### Diff
4
+
5
+ ```diff
6
+ +++ b/src/services/cost-tracker.ts
7
+ @@ -0,0 +1,22 @@
8
+ +export class CostTracker {
9
+ + async trackRun(runId: string, model: string, output: string): Promise<CostRecord> {
10
+ + // Estimate tokens from output length
11
+ + const estimatedTokens = Math.ceil(output.length / 4);
12
+ + const estimatedInputTokens = estimatedTokens * 2;
13
+ +
14
+ + const cost = this.calculateCost(model, estimatedInputTokens, estimatedTokens);
15
+ +
16
+ + return {
17
+ + runId,
18
+ + model,
19
+ + inputTokens: estimatedInputTokens,
20
+ + outputTokens: estimatedTokens,
21
+ + costUsd: cost,
22
+ + };
23
+ + }
24
+ +
25
+ + private calculateCost(model: string, input: number, output: number): number {
26
+ + const rates: Record<string, number> = { 'gpt-4': 0.03, 'claude': 0.015 };
27
+ + return (input + output) * (rates[model] ?? 0.01) / 1000;
28
+ + }
29
+ +}
30
+
31
+ +++ b/src/components/CostDashboard.tsx
32
+ @@ -0,0 +1,15 @@
33
+ +export function CostDashboard({ runs }: { runs: CostRecord[] }) {
34
+ + const totalCost = runs.reduce((sum, r) => sum + r.costUsd, 0);
35
+ + return (
36
+ + <div>
37
+ + <h2>Cost Summary</h2>
38
+ + <p>Total: ${totalCost.toFixed(2)}</p>
39
+ + {runs.map(r => (
40
+ + <div key={r.runId}>
41
+ + <span>{r.model}: {r.inputTokens} in / {r.outputTokens} out</span>
42
+ + <span>${r.costUsd.toFixed(4)}</span>
43
+ + </div>
44
+ + ))}
45
+ + </div>
46
+ + );
47
+ +}
48
+ ```
49
+
50
+ ### Analysis Required
51
+
52
+ Review this diff. The issue asked for cost tracking of pipeline runs.
@@ -0,0 +1,10 @@
1
+ id: 010-fabricated-metrics
2
+ description: Review should flag hardcoded or estimated values displayed as real metrics
3
+ type: step
4
+ step: review
5
+ tags:
6
+ - wiring
7
+ - metrics
8
+ - quality-failure
9
+ source: distribution
10
+ timeout: 60