@tangle-network/agent-bench 0.3.8 → 0.4.1

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 (115) hide show
  1. package/CHANGELOG.md +13 -0
  2. package/README.md +7 -0
  3. package/dist/adapters.js +23 -23
  4. package/dist/benchmarks/_harness.d.ts +4 -1
  5. package/dist/benchmarks/_harness.js +3 -1
  6. package/dist/benchmarks/aec-bench.js +2 -2
  7. package/dist/benchmarks/agentbench.js +2 -2
  8. package/dist/benchmarks/appworld.js +2 -2
  9. package/dist/benchmarks/bfcl.js +2 -2
  10. package/dist/benchmarks/commit0.js +2 -2
  11. package/dist/benchmarks/crag.js +2 -2
  12. package/dist/benchmarks/dabstep.js +2 -2
  13. package/dist/benchmarks/enterpriseops-gym.js +2 -2
  14. package/dist/benchmarks/finresearchbench.js +2 -2
  15. package/dist/benchmarks/nomiracl.js +2 -2
  16. package/dist/benchmarks/open-rag-bench.js +2 -2
  17. package/dist/benchmarks/programbench.js +2 -2
  18. package/dist/benchmarks/ragbench.js +2 -2
  19. package/dist/benchmarks/swe-bench.js +2 -2
  20. package/dist/benchmarks/t2-ragbench.js +2 -2
  21. package/dist/benchmarks/tau-bench-shared.js +2 -2
  22. package/dist/benchmarks/tau2-bench.js +3 -3
  23. package/dist/benchmarks/tau3-banking.js +3 -3
  24. package/dist/benchmarks/terminal-bench.js +2 -2
  25. package/dist/benchmarks/toollm.js +2 -2
  26. package/dist/benchmarks/webarena-verified.js +2 -2
  27. package/dist/{chunk-YSMEKBTD.js → chunk-5FEQDSCT.js} +2 -2
  28. package/dist/{chunk-PB64GYIG.js → chunk-67ACKDCX.js} +2 -2
  29. package/dist/{chunk-NQG5XDSB.js → chunk-7FKBWOQT.js} +2 -2
  30. package/dist/{chunk-K3BQGZCT.js → chunk-BEN6IF2X.js} +2 -2
  31. package/dist/{chunk-SHYIRB7I.js → chunk-BZY5QARD.js} +2 -2
  32. package/dist/{chunk-KP5KD6EN.js → chunk-CLIKAXKH.js} +2 -2
  33. package/dist/{chunk-XKEFIFIC.js → chunk-CWIOBFSP.js} +2 -2
  34. package/dist/{chunk-IA2FBTWC.js → chunk-CXDUTWQE.js} +2 -2
  35. package/dist/{chunk-RH5F53JT.js → chunk-DWALFME7.js} +2 -2
  36. package/dist/{chunk-V7AEBY6U.js → chunk-EEOC6QPJ.js} +21 -21
  37. package/dist/{chunk-WSKWVEQB.js → chunk-EIETHPD5.js} +7 -3
  38. package/dist/chunk-EIETHPD5.js.map +1 -0
  39. package/dist/{chunk-SHM6MRRF.js → chunk-GC2EPS6L.js} +2 -2
  40. package/dist/{chunk-IZ5M6OAC.js → chunk-GCHL6XPM.js} +2 -2
  41. package/dist/{chunk-IFVINJ4B.js → chunk-HQ5HCCKF.js} +2 -2
  42. package/dist/{chunk-7GRVHU22.js → chunk-HVW25KSX.js} +2 -2
  43. package/dist/{chunk-3U5TXJZS.js → chunk-J6BU3NTM.js} +2 -2
  44. package/dist/{chunk-XYA4XSNU.js → chunk-JSQOUKXS.js} +4 -3
  45. package/dist/chunk-JSQOUKXS.js.map +1 -0
  46. package/dist/{chunk-SVR2LKYI.js → chunk-NRMGT25X.js} +2 -2
  47. package/dist/{chunk-MQMRLGOG.js → chunk-QZZEAHWJ.js} +2 -2
  48. package/dist/{chunk-RCYQEFNX.js → chunk-WG7TM7UV.js} +3 -3
  49. package/dist/{chunk-HWST3SED.js → chunk-XXFF3RRD.js} +2 -2
  50. package/dist/{chunk-SFLA7OH3.js → chunk-ZFNOM7WR.js} +3 -3
  51. package/dist/{chunk-Z4TZ76N7.js → chunk-ZNCCYTFG.js} +2 -2
  52. package/dist/index.js +23 -23
  53. package/package.json +4 -4
  54. package/pier_agents/candidate_contract.py +238 -24
  55. package/pier_agents/tangle_candidate.py +75 -5
  56. package/scripts/trata-hedge/README.md +6 -5
  57. package/scripts/verify-packed-consumer.mjs +65 -1
  58. package/scripts/verify-pier-agent.mts +6 -4
  59. package/src/benchmarks/_harness.test.mts +16 -1
  60. package/src/benchmarks/_harness.ts +9 -2
  61. package/src/benchmarks/terminal-bench.ts +2 -1
  62. package/src/gate.ts +1 -1
  63. package/src/hev-eval.mts +5 -2
  64. package/src/hev-improve.mts +118 -73
  65. package/src/official-optimizer-config.mts +89 -0
  66. package/src/official-optimizer-config.test.mts +88 -0
  67. package/src/profiles.ts +2 -2
  68. package/src/rollout-ledger/backfill-swe-arena.test.mts +28 -24
  69. package/src/smoke-structural-rollout.mts +15 -9
  70. package/src/swe-arena/activation.mts +1 -4
  71. package/src/swe-arena/activation.test.mts +10 -13
  72. package/src/swe-arena/gepa-seat.mts +425 -131
  73. package/src/swe-arena/gepa-seat.test.mts +524 -100
  74. package/src/swe-arena/implementation-ref.test.mts +64 -0
  75. package/src/swe-arena/implementation-ref.ts +62 -0
  76. package/src/swe-arena/outer-loop.mts +103 -76
  77. package/src/swe-arena/proposer-fanout.mts +51 -36
  78. package/src/swe-arena/proposer-fanout.test.mts +0 -1
  79. package/src/swe-arena/proposer-provenance.mts +11 -16
  80. package/src/swe-arena/scratch-worktree.test.mts +55 -0
  81. package/src/swe-arena/scratch-worktree.ts +34 -0
  82. package/src/swe-code-improve.mts +24 -25
  83. package/src/swe-improve.mts +129 -96
  84. package/src/swe-local-proof.mts +6 -1
  85. package/src/swe-stream.mts +4 -2
  86. package/src/tb-container-executor.test.mts +30 -6
  87. package/src/tb-supervisor-sidecar.mts +2 -1
  88. package/src/trata-gepa.mts +182 -245
  89. package/dist/chunk-WSKWVEQB.js.map +0 -1
  90. package/dist/chunk-XYA4XSNU.js.map +0 -1
  91. package/src/live-improve-campaign-mbpp.mts +0 -641
  92. package/src/live-improve-campaign.mts +0 -500
  93. package/src/swe-arena/lineage-record.mts +0 -164
  94. package/src/swe-arena/lineage-record.test.mts +0 -115
  95. /package/dist/{chunk-YSMEKBTD.js.map → chunk-5FEQDSCT.js.map} +0 -0
  96. /package/dist/{chunk-PB64GYIG.js.map → chunk-67ACKDCX.js.map} +0 -0
  97. /package/dist/{chunk-NQG5XDSB.js.map → chunk-7FKBWOQT.js.map} +0 -0
  98. /package/dist/{chunk-K3BQGZCT.js.map → chunk-BEN6IF2X.js.map} +0 -0
  99. /package/dist/{chunk-SHYIRB7I.js.map → chunk-BZY5QARD.js.map} +0 -0
  100. /package/dist/{chunk-KP5KD6EN.js.map → chunk-CLIKAXKH.js.map} +0 -0
  101. /package/dist/{chunk-XKEFIFIC.js.map → chunk-CWIOBFSP.js.map} +0 -0
  102. /package/dist/{chunk-IA2FBTWC.js.map → chunk-CXDUTWQE.js.map} +0 -0
  103. /package/dist/{chunk-RH5F53JT.js.map → chunk-DWALFME7.js.map} +0 -0
  104. /package/dist/{chunk-V7AEBY6U.js.map → chunk-EEOC6QPJ.js.map} +0 -0
  105. /package/dist/{chunk-SHM6MRRF.js.map → chunk-GC2EPS6L.js.map} +0 -0
  106. /package/dist/{chunk-IZ5M6OAC.js.map → chunk-GCHL6XPM.js.map} +0 -0
  107. /package/dist/{chunk-IFVINJ4B.js.map → chunk-HQ5HCCKF.js.map} +0 -0
  108. /package/dist/{chunk-7GRVHU22.js.map → chunk-HVW25KSX.js.map} +0 -0
  109. /package/dist/{chunk-3U5TXJZS.js.map → chunk-J6BU3NTM.js.map} +0 -0
  110. /package/dist/{chunk-SVR2LKYI.js.map → chunk-NRMGT25X.js.map} +0 -0
  111. /package/dist/{chunk-MQMRLGOG.js.map → chunk-QZZEAHWJ.js.map} +0 -0
  112. /package/dist/{chunk-RCYQEFNX.js.map → chunk-WG7TM7UV.js.map} +0 -0
  113. /package/dist/{chunk-HWST3SED.js.map → chunk-XXFF3RRD.js.map} +0 -0
  114. /package/dist/{chunk-SFLA7OH3.js.map → chunk-ZFNOM7WR.js.map} +0 -0
  115. /package/dist/{chunk-Z4TZ76N7.js.map → chunk-ZNCCYTFG.js.map} +0 -0
package/CHANGELOG.md CHANGED
@@ -1,5 +1,18 @@
1
1
  # Changelog
2
2
 
3
+ ## 0.4.1
4
+
5
+ - Read Runtime 0.105 candidate plans from their signed run cell, benchmark records, and profile activation.
6
+ - Allow only the fixed public executable path signed by the Runtime plan.
7
+ - Restore the real Pier failure/success proof against the current receipt layout.
8
+
9
+ ## 0.4.0
10
+
11
+ - Add the resumable SWE improvement loop backed by the official GEPA engine.
12
+ - Isolate every concurrent candidate in its own Git worktree and authorize the exact candidate before execution.
13
+ - Include the complete Runtime, Bench, and installed SWE judge implementations in resume identity.
14
+ - Consume `@tangle-network/agent-runtime` 0.105.x, `@tangle-network/agent-eval` 0.126.6, and `@tangle-network/agent-knowledge` 5.0.1.
15
+
3
16
  ## 0.3.8
4
17
 
5
18
  - Resolve package-owned fixtures and scripts from both source and compiled installs, and prove ToolLLM fixture loading from the packed package.
package/README.md CHANGED
@@ -26,6 +26,13 @@ pnpm install # tsx + link parent
26
26
 
27
27
  The judge needs only Docker; workers need a model key (Tangle router `TANGLE_API_KEY`, or a direct provider).
28
28
 
29
+ Live optimizer scripts require explicit token prices so cost records cannot be guessed.
30
+ Use the `REFLECT_` prefix for the general benchmark scripts and `GEPA_OPTIMIZER_` for the SWE arena seat.
31
+ For either prefix, set `INPUT_USD_PER_MILLION`, `CACHED_INPUT_USD_PER_MILLION`, `CACHE_WRITE_USD_PER_MILLION`, and `OUTPUT_USD_PER_MILLION`.
32
+ The SWE arena seat reads its model from `GEPA_OPTIMIZER_MODEL`, its URL from `GEPA_OPTIMIZER_BASE_URL`, and its key from `GEPA_OPTIMIZER_API_KEY`.
33
+ It falls back to the configured driver model, `ROUTER_BASE`, and `TANGLE_API_KEY`.
34
+ Missing prices fail before an optimizer model call.
35
+
29
36
  Retain every official per-test log and report before the temporary evaluator directory is removed:
30
37
 
31
38
  ```ts
package/dist/adapters.js CHANGED
@@ -1,40 +1,40 @@
1
1
  import {
2
2
  ADAPTERS,
3
3
  resolveAdapter
4
- } from "./chunk-V7AEBY6U.js";
5
- import "./chunk-XKEFIFIC.js";
4
+ } from "./chunk-EEOC6QPJ.js";
5
+ import "./chunk-CWIOBFSP.js";
6
6
  import "./chunk-VQRS7VUC.js";
7
- import "./chunk-HWST3SED.js";
8
- import "./chunk-IFVINJ4B.js";
7
+ import "./chunk-XXFF3RRD.js";
8
+ import "./chunk-HQ5HCCKF.js";
9
9
  import "./chunk-SEVJPLZC.js";
10
- import "./chunk-3U5TXJZS.js";
11
- import "./chunk-SHM6MRRF.js";
12
- import "./chunk-RCYQEFNX.js";
13
- import "./chunk-SFLA7OH3.js";
14
- import "./chunk-IZ5M6OAC.js";
15
- import "./chunk-XYA4XSNU.js";
10
+ import "./chunk-J6BU3NTM.js";
11
+ import "./chunk-GC2EPS6L.js";
12
+ import "./chunk-WG7TM7UV.js";
13
+ import "./chunk-ZFNOM7WR.js";
14
+ import "./chunk-GCHL6XPM.js";
15
+ import "./chunk-JSQOUKXS.js";
16
16
  import "./chunk-INNOYXCP.js";
17
17
  import "./chunk-PA2ZKHJC.js";
18
18
  import "./chunk-5H5XV76F.js";
19
19
  import "./chunk-TBKU5XQI.js";
20
- import "./chunk-NQG5XDSB.js";
21
- import "./chunk-SHYIRB7I.js";
22
- import "./chunk-MQMRLGOG.js";
20
+ import "./chunk-7FKBWOQT.js";
21
+ import "./chunk-BZY5QARD.js";
22
+ import "./chunk-QZZEAHWJ.js";
23
23
  import "./chunk-PUIRNYI7.js";
24
24
  import "./chunk-C7T7WEK2.js";
25
25
  import "./chunk-JTHWEDEW.js";
26
- import "./chunk-Z4TZ76N7.js";
27
- import "./chunk-SVR2LKYI.js";
26
+ import "./chunk-ZNCCYTFG.js";
27
+ import "./chunk-NRMGT25X.js";
28
28
  import "./chunk-X3BTXCJ4.js";
29
- import "./chunk-RH5F53JT.js";
30
- import "./chunk-YSMEKBTD.js";
31
- import "./chunk-7GRVHU22.js";
29
+ import "./chunk-DWALFME7.js";
30
+ import "./chunk-5FEQDSCT.js";
31
+ import "./chunk-HVW25KSX.js";
32
32
  import "./chunk-UPAMRDX4.js";
33
- import "./chunk-K3BQGZCT.js";
34
- import "./chunk-PB64GYIG.js";
35
- import "./chunk-IA2FBTWC.js";
36
- import "./chunk-KP5KD6EN.js";
37
- import "./chunk-WSKWVEQB.js";
33
+ import "./chunk-BEN6IF2X.js";
34
+ import "./chunk-67ACKDCX.js";
35
+ import "./chunk-CXDUTWQE.js";
36
+ import "./chunk-CLIKAXKH.js";
37
+ import "./chunk-EIETHPD5.js";
38
38
  import "./chunk-53UPUNBZ.js";
39
39
  export {
40
40
  ADAPTERS,
@@ -31,6 +31,9 @@ declare const venvPython: string;
31
31
  * deps conflict with the shared `.venv` (commit0 downgrades pydantic/sqlalchemy)
32
32
  * get their own venv and pass its python explicitly — keeping the shared one clean. */
33
33
  declare const venvPythonAt: (venvDir: string) => string;
34
+ /** Resolve an executable in an isolated venv. Relative venv paths are package-owned;
35
+ * absolute paths allow installed consumers to keep large environments elsewhere. */
36
+ declare function venvBinAt(venvDir: string, name: string): string;
34
37
  /** Report/transcript reads are large; 256 MiB matches the SWE harness budget. */
35
38
  declare const bigBuffer: number;
36
39
  /** Path to a named executable inside the bench venv (e.g. `venvBin('tb')`). */
@@ -122,4 +125,4 @@ declare function stageFile(path: string, content: string): Promise<void>;
122
125
  /** Sanitize an instance id into a filesystem/run-id-safe token. */
123
126
  declare function safeRunId(prefix: string, id: string): string;
124
127
 
125
- export { StagedJudgeError, type StagedRunCaptureSpec, type StagedRunSpec, benchRoot, bigBuffer, preflightVenvImports, readJsonReport, resolveBenchPython, runStagedJudge, runVenvPython, runVenvScriptStdin, safeRunId, stageFile, venvBin, venvPython, venvPythonAt };
128
+ export { StagedJudgeError, type StagedRunCaptureSpec, type StagedRunSpec, benchRoot, bigBuffer, preflightVenvImports, readJsonReport, resolveBenchPython, runStagedJudge, runVenvPython, runVenvScriptStdin, safeRunId, stageFile, venvBin, venvBinAt, venvPython, venvPythonAt };
@@ -11,9 +11,10 @@ import {
11
11
  safeRunId,
12
12
  stageFile,
13
13
  venvBin,
14
+ venvBinAt,
14
15
  venvPython,
15
16
  venvPythonAt
16
- } from "../chunk-WSKWVEQB.js";
17
+ } from "../chunk-EIETHPD5.js";
17
18
  export {
18
19
  StagedJudgeError,
19
20
  benchRoot,
@@ -27,6 +28,7 @@ export {
27
28
  safeRunId,
28
29
  stageFile,
29
30
  venvBin,
31
+ venvBinAt,
30
32
  venvPython,
31
33
  venvPythonAt
32
34
  };
@@ -1,7 +1,7 @@
1
1
  import {
2
2
  createAecBenchAdapter
3
- } from "../chunk-K3BQGZCT.js";
4
- import "../chunk-WSKWVEQB.js";
3
+ } from "../chunk-BEN6IF2X.js";
4
+ import "../chunk-EIETHPD5.js";
5
5
  export {
6
6
  createAecBenchAdapter
7
7
  };
@@ -1,8 +1,8 @@
1
1
  import {
2
2
  agentbenchAnswerOutput,
3
3
  createAgentBenchAdapter
4
- } from "../chunk-PB64GYIG.js";
5
- import "../chunk-WSKWVEQB.js";
4
+ } from "../chunk-67ACKDCX.js";
5
+ import "../chunk-EIETHPD5.js";
6
6
  export {
7
7
  agentbenchAnswerOutput,
8
8
  createAgentBenchAdapter
@@ -3,8 +3,8 @@ import {
3
3
  appworldToolLoopClient,
4
4
  createAppWorldAdapter,
5
5
  createAppWorldReactAdapter
6
- } from "../chunk-IA2FBTWC.js";
7
- import "../chunk-WSKWVEQB.js";
6
+ } from "../chunk-CXDUTWQE.js";
7
+ import "../chunk-EIETHPD5.js";
8
8
  export {
9
9
  appworldSolutionOutput,
10
10
  appworldToolLoopClient,
@@ -1,8 +1,8 @@
1
1
  import {
2
2
  bfclOutput,
3
3
  createBfclAdapter
4
- } from "../chunk-KP5KD6EN.js";
5
- import "../chunk-WSKWVEQB.js";
4
+ } from "../chunk-CLIKAXKH.js";
5
+ import "../chunk-EIETHPD5.js";
6
6
  export {
7
7
  bfclOutput,
8
8
  createBfclAdapter
@@ -1,8 +1,8 @@
1
1
  import {
2
2
  commit0DiffOutput,
3
3
  createCommit0Adapter
4
- } from "../chunk-Z4TZ76N7.js";
5
- import "../chunk-WSKWVEQB.js";
4
+ } from "../chunk-ZNCCYTFG.js";
5
+ import "../chunk-EIETHPD5.js";
6
6
  export {
7
7
  commit0DiffOutput,
8
8
  createCommit0Adapter
@@ -1,8 +1,8 @@
1
1
  import {
2
2
  createCragAdapter
3
- } from "../chunk-SVR2LKYI.js";
3
+ } from "../chunk-NRMGT25X.js";
4
4
  import "../chunk-X3BTXCJ4.js";
5
- import "../chunk-WSKWVEQB.js";
5
+ import "../chunk-EIETHPD5.js";
6
6
  export {
7
7
  createCragAdapter
8
8
  };
@@ -1,8 +1,8 @@
1
1
  import {
2
2
  createDabstepAdapter,
3
3
  dabstepAnswerOutput
4
- } from "../chunk-RH5F53JT.js";
5
- import "../chunk-WSKWVEQB.js";
4
+ } from "../chunk-DWALFME7.js";
5
+ import "../chunk-EIETHPD5.js";
6
6
  export {
7
7
  createDabstepAdapter,
8
8
  dabstepAnswerOutput
@@ -1,8 +1,8 @@
1
1
  import {
2
2
  createEnterpriseOpsGymAdapter,
3
3
  enterpriseOpsTranscriptOutput
4
- } from "../chunk-YSMEKBTD.js";
5
- import "../chunk-WSKWVEQB.js";
4
+ } from "../chunk-5FEQDSCT.js";
5
+ import "../chunk-EIETHPD5.js";
6
6
  export {
7
7
  createEnterpriseOpsGymAdapter,
8
8
  enterpriseOpsTranscriptOutput
@@ -1,7 +1,7 @@
1
1
  import {
2
2
  createFinResearchBenchAdapter
3
- } from "../chunk-7GRVHU22.js";
4
- import "../chunk-WSKWVEQB.js";
3
+ } from "../chunk-HVW25KSX.js";
4
+ import "../chunk-EIETHPD5.js";
5
5
  export {
6
6
  createFinResearchBenchAdapter
7
7
  };
@@ -1,8 +1,8 @@
1
1
  import {
2
2
  createNoMiraclAdapter
3
- } from "../chunk-NQG5XDSB.js";
3
+ } from "../chunk-7FKBWOQT.js";
4
4
  import "../chunk-X3BTXCJ4.js";
5
- import "../chunk-WSKWVEQB.js";
5
+ import "../chunk-EIETHPD5.js";
6
6
  export {
7
7
  createNoMiraclAdapter
8
8
  };
@@ -1,8 +1,8 @@
1
1
  import {
2
2
  createOpenRagBenchAdapter
3
- } from "../chunk-SHYIRB7I.js";
3
+ } from "../chunk-BZY5QARD.js";
4
4
  import "../chunk-X3BTXCJ4.js";
5
- import "../chunk-WSKWVEQB.js";
5
+ import "../chunk-EIETHPD5.js";
6
6
  export {
7
7
  createOpenRagBenchAdapter
8
8
  };
@@ -1,8 +1,8 @@
1
1
  import {
2
2
  createProgrambenchAdapter,
3
3
  programbenchSubmissionOutput
4
- } from "../chunk-MQMRLGOG.js";
5
- import "../chunk-WSKWVEQB.js";
4
+ } from "../chunk-QZZEAHWJ.js";
5
+ import "../chunk-EIETHPD5.js";
6
6
  export {
7
7
  createProgrambenchAdapter,
8
8
  programbenchSubmissionOutput
@@ -1,8 +1,8 @@
1
1
  import {
2
2
  createRagBenchAdapter
3
- } from "../chunk-IFVINJ4B.js";
3
+ } from "../chunk-HQ5HCCKF.js";
4
4
  import "../chunk-X3BTXCJ4.js";
5
- import "../chunk-WSKWVEQB.js";
5
+ import "../chunk-EIETHPD5.js";
6
6
  export {
7
7
  createRagBenchAdapter
8
8
  };
@@ -3,8 +3,8 @@ import {
3
3
  scoreSweReport,
4
4
  sweEvaluationArgv,
5
5
  swePatchOutput
6
- } from "../chunk-3U5TXJZS.js";
7
- import "../chunk-WSKWVEQB.js";
6
+ } from "../chunk-J6BU3NTM.js";
7
+ import "../chunk-EIETHPD5.js";
8
8
  export {
9
9
  createSweBenchAdapter,
10
10
  scoreSweReport,
@@ -1,8 +1,8 @@
1
1
  import {
2
2
  createT2RagBenchAdapter
3
- } from "../chunk-SHM6MRRF.js";
3
+ } from "../chunk-GC2EPS6L.js";
4
4
  import "../chunk-X3BTXCJ4.js";
5
- import "../chunk-WSKWVEQB.js";
5
+ import "../chunk-EIETHPD5.js";
6
6
  export {
7
7
  createT2RagBenchAdapter
8
8
  };
@@ -1,8 +1,8 @@
1
1
  import {
2
2
  createTauBenchAdapter,
3
3
  tauResultsOutput
4
- } from "../chunk-IZ5M6OAC.js";
5
- import "../chunk-WSKWVEQB.js";
4
+ } from "../chunk-GCHL6XPM.js";
5
+ import "../chunk-EIETHPD5.js";
6
6
  export {
7
7
  createTauBenchAdapter,
8
8
  tauResultsOutput
@@ -1,9 +1,9 @@
1
1
  import {
2
2
  createTau2BenchAdapter,
3
3
  tau2ResultsOutput
4
- } from "../chunk-RCYQEFNX.js";
5
- import "../chunk-IZ5M6OAC.js";
6
- import "../chunk-WSKWVEQB.js";
4
+ } from "../chunk-WG7TM7UV.js";
5
+ import "../chunk-GCHL6XPM.js";
6
+ import "../chunk-EIETHPD5.js";
7
7
  export {
8
8
  createTau2BenchAdapter,
9
9
  tau2ResultsOutput
@@ -1,8 +1,8 @@
1
1
  import {
2
2
  createTau3BankingAdapter
3
- } from "../chunk-SFLA7OH3.js";
4
- import "../chunk-IZ5M6OAC.js";
5
- import "../chunk-WSKWVEQB.js";
3
+ } from "../chunk-ZFNOM7WR.js";
4
+ import "../chunk-GCHL6XPM.js";
5
+ import "../chunk-EIETHPD5.js";
6
6
  export {
7
7
  createTau3BankingAdapter
8
8
  };
@@ -1,7 +1,7 @@
1
1
  import {
2
2
  createTerminalBenchAdapter
3
- } from "../chunk-XYA4XSNU.js";
4
- import "../chunk-WSKWVEQB.js";
3
+ } from "../chunk-JSQOUKXS.js";
4
+ import "../chunk-EIETHPD5.js";
5
5
  export {
6
6
  createTerminalBenchAdapter
7
7
  };
@@ -1,8 +1,8 @@
1
1
  import {
2
2
  createToolLlmAdapter,
3
3
  toollmOutput
4
- } from "../chunk-XKEFIFIC.js";
5
- import "../chunk-WSKWVEQB.js";
4
+ } from "../chunk-CWIOBFSP.js";
5
+ import "../chunk-EIETHPD5.js";
6
6
  export {
7
7
  createToolLlmAdapter,
8
8
  toollmOutput
@@ -1,8 +1,8 @@
1
1
  import {
2
2
  createWebArenaVerifiedAdapter,
3
3
  webarenaOutputDirOutput
4
- } from "../chunk-HWST3SED.js";
5
- import "../chunk-WSKWVEQB.js";
4
+ } from "../chunk-XXFF3RRD.js";
5
+ import "../chunk-EIETHPD5.js";
6
6
  export {
7
7
  createWebArenaVerifiedAdapter,
8
8
  webarenaOutputDirOutput
@@ -1,7 +1,7 @@
1
1
  import {
2
2
  benchRoot,
3
3
  runVenvScriptStdin
4
- } from "./chunk-WSKWVEQB.js";
4
+ } from "./chunk-EIETHPD5.js";
5
5
 
6
6
  // src/benchmarks/enterpriseops-gym.ts
7
7
  import { mkdir, readFile, rm, writeFile } from "fs/promises";
@@ -208,4 +208,4 @@ export {
208
208
  enterpriseOpsTranscriptOutput,
209
209
  createEnterpriseOpsGymAdapter
210
210
  };
211
- //# sourceMappingURL=chunk-YSMEKBTD.js.map
211
+ //# sourceMappingURL=chunk-5FEQDSCT.js.map
@@ -1,6 +1,6 @@
1
1
  import {
2
2
  benchRoot
3
- } from "./chunk-WSKWVEQB.js";
3
+ } from "./chunk-EIETHPD5.js";
4
4
 
5
5
  // src/benchmarks/agentbench.ts
6
6
  import { readFile } from "fs/promises";
@@ -115,4 +115,4 @@ export {
115
115
  agentbenchAnswerOutput,
116
116
  createAgentBenchAdapter
117
117
  };
118
- //# sourceMappingURL=chunk-PB64GYIG.js.map
118
+ //# sourceMappingURL=chunk-67ACKDCX.js.map
@@ -10,7 +10,7 @@ import {
10
10
  } from "./chunk-X3BTXCJ4.js";
11
11
  import {
12
12
  benchRoot
13
- } from "./chunk-WSKWVEQB.js";
13
+ } from "./chunk-EIETHPD5.js";
14
14
 
15
15
  // src/benchmarks/nomiracl.ts
16
16
  import { readFile } from "fs/promises";
@@ -144,4 +144,4 @@ function createNoMiraclAdapter() {
144
144
  export {
145
145
  createNoMiraclAdapter
146
146
  };
147
- //# sourceMappingURL=chunk-NQG5XDSB.js.map
147
+ //# sourceMappingURL=chunk-7FKBWOQT.js.map
@@ -5,7 +5,7 @@ import {
5
5
  runStagedJudge,
6
6
  stageFile,
7
7
  venvPython
8
- } from "./chunk-WSKWVEQB.js";
8
+ } from "./chunk-EIETHPD5.js";
9
9
 
10
10
  // src/benchmarks/aec-bench.ts
11
11
  import { readFile } from "fs/promises";
@@ -218,4 +218,4 @@ Fix: ensure network access to raw.githubusercontent.com, or set AEC_FIXTURES=1 t
218
218
  export {
219
219
  createAecBenchAdapter
220
220
  };
221
- //# sourceMappingURL=chunk-K3BQGZCT.js.map
221
+ //# sourceMappingURL=chunk-BEN6IF2X.js.map
@@ -14,7 +14,7 @@ import {
14
14
  } from "./chunk-X3BTXCJ4.js";
15
15
  import {
16
16
  benchRoot
17
- } from "./chunk-WSKWVEQB.js";
17
+ } from "./chunk-EIETHPD5.js";
18
18
 
19
19
  // src/benchmarks/open-rag-bench.ts
20
20
  import { readFile } from "fs/promises";
@@ -117,4 +117,4 @@ function createOpenRagBenchAdapter() {
117
117
  export {
118
118
  createOpenRagBenchAdapter
119
119
  };
120
- //# sourceMappingURL=chunk-SHYIRB7I.js.map
120
+ //# sourceMappingURL=chunk-BZY5QARD.js.map
@@ -1,6 +1,6 @@
1
1
  import {
2
2
  benchRoot
3
- } from "./chunk-WSKWVEQB.js";
3
+ } from "./chunk-EIETHPD5.js";
4
4
 
5
5
  // src/benchmarks/bfcl.ts
6
6
  import { readFile, stat } from "fs/promises";
@@ -273,4 +273,4 @@ export {
273
273
  bfclOutput,
274
274
  createBfclAdapter
275
275
  };
276
- //# sourceMappingURL=chunk-KP5KD6EN.js.map
276
+ //# sourceMappingURL=chunk-CLIKAXKH.js.map
@@ -1,6 +1,6 @@
1
1
  import {
2
2
  benchRoot
3
- } from "./chunk-WSKWVEQB.js";
3
+ } from "./chunk-EIETHPD5.js";
4
4
 
5
5
  // src/benchmarks/toollm.ts
6
6
  import { readFile } from "fs/promises";
@@ -194,4 +194,4 @@ export {
194
194
  toollmOutput,
195
195
  createToolLlmAdapter
196
196
  };
197
- //# sourceMappingURL=chunk-XKEFIFIC.js.map
197
+ //# sourceMappingURL=chunk-CWIOBFSP.js.map
@@ -3,7 +3,7 @@ import {
3
3
  preflightVenvImports,
4
4
  runVenvScriptStdin,
5
5
  venvPython
6
- } from "./chunk-WSKWVEQB.js";
6
+ } from "./chunk-EIETHPD5.js";
7
7
 
8
8
  // src/benchmarks/appworld.ts
9
9
  import { spawn } from "child_process";
@@ -315,4 +315,4 @@ export {
315
315
  appworldToolLoopClient,
316
316
  createAppWorldReactAdapter
317
317
  };
318
- //# sourceMappingURL=chunk-IA2FBTWC.js.map
318
+ //# sourceMappingURL=chunk-CXDUTWQE.js.map
@@ -2,7 +2,7 @@ import {
2
2
  benchRoot,
3
3
  runVenvPython,
4
4
  runVenvScriptStdin
5
- } from "./chunk-WSKWVEQB.js";
5
+ } from "./chunk-EIETHPD5.js";
6
6
 
7
7
  // src/benchmarks/dabstep.ts
8
8
  import { join } from "path";
@@ -179,4 +179,4 @@ export {
179
179
  dabstepAnswerOutput,
180
180
  createDabstepAdapter
181
181
  };
182
- //# sourceMappingURL=chunk-RH5F53JT.js.map
182
+ //# sourceMappingURL=chunk-DWALFME7.js.map
@@ -1,33 +1,33 @@
1
1
  import {
2
2
  createToolLlmAdapter
3
- } from "./chunk-XKEFIFIC.js";
3
+ } from "./chunk-CWIOBFSP.js";
4
4
  import {
5
5
  createTrataHedgeAdapter
6
6
  } from "./chunk-VQRS7VUC.js";
7
7
  import {
8
8
  createWebArenaVerifiedAdapter
9
- } from "./chunk-HWST3SED.js";
9
+ } from "./chunk-XXFF3RRD.js";
10
10
  import {
11
11
  createRagBenchAdapter
12
- } from "./chunk-IFVINJ4B.js";
12
+ } from "./chunk-HQ5HCCKF.js";
13
13
  import {
14
14
  createSimpleQaAdapter
15
15
  } from "./chunk-SEVJPLZC.js";
16
16
  import {
17
17
  createSweBenchAdapter
18
- } from "./chunk-3U5TXJZS.js";
18
+ } from "./chunk-J6BU3NTM.js";
19
19
  import {
20
20
  createT2RagBenchAdapter
21
- } from "./chunk-SHM6MRRF.js";
21
+ } from "./chunk-GC2EPS6L.js";
22
22
  import {
23
23
  createTau2BenchAdapter
24
- } from "./chunk-RCYQEFNX.js";
24
+ } from "./chunk-WG7TM7UV.js";
25
25
  import {
26
26
  createTau3BankingAdapter
27
- } from "./chunk-SFLA7OH3.js";
27
+ } from "./chunk-ZFNOM7WR.js";
28
28
  import {
29
29
  createTerminalBenchAdapter
30
- } from "./chunk-XYA4XSNU.js";
30
+ } from "./chunk-JSQOUKXS.js";
31
31
  import {
32
32
  createFramesAdapter
33
33
  } from "./chunk-INNOYXCP.js";
@@ -42,13 +42,13 @@ import {
42
42
  } from "./chunk-TBKU5XQI.js";
43
43
  import {
44
44
  createNoMiraclAdapter
45
- } from "./chunk-NQG5XDSB.js";
45
+ } from "./chunk-7FKBWOQT.js";
46
46
  import {
47
47
  createOpenRagBenchAdapter
48
- } from "./chunk-SHYIRB7I.js";
48
+ } from "./chunk-BZY5QARD.js";
49
49
  import {
50
50
  createProgrambenchAdapter
51
- } from "./chunk-MQMRLGOG.js";
51
+ } from "./chunk-QZZEAHWJ.js";
52
52
  import {
53
53
  createCadBenchAdapter
54
54
  } from "./chunk-PUIRNYI7.js";
@@ -57,35 +57,35 @@ import {
57
57
  } from "./chunk-C7T7WEK2.js";
58
58
  import {
59
59
  createCommit0Adapter
60
- } from "./chunk-Z4TZ76N7.js";
60
+ } from "./chunk-ZNCCYTFG.js";
61
61
  import {
62
62
  createCragAdapter
63
- } from "./chunk-SVR2LKYI.js";
63
+ } from "./chunk-NRMGT25X.js";
64
64
  import {
65
65
  createDabstepAdapter
66
- } from "./chunk-RH5F53JT.js";
66
+ } from "./chunk-DWALFME7.js";
67
67
  import {
68
68
  createEnterpriseOpsGymAdapter
69
- } from "./chunk-YSMEKBTD.js";
69
+ } from "./chunk-5FEQDSCT.js";
70
70
  import {
71
71
  createFinResearchBenchAdapter
72
- } from "./chunk-7GRVHU22.js";
72
+ } from "./chunk-HVW25KSX.js";
73
73
  import {
74
74
  createFinsearchcompAdapter
75
75
  } from "./chunk-UPAMRDX4.js";
76
76
  import {
77
77
  createAecBenchAdapter
78
- } from "./chunk-K3BQGZCT.js";
78
+ } from "./chunk-BEN6IF2X.js";
79
79
  import {
80
80
  createAgentBenchAdapter
81
- } from "./chunk-PB64GYIG.js";
81
+ } from "./chunk-67ACKDCX.js";
82
82
  import {
83
83
  createAppWorldAdapter,
84
84
  createAppWorldReactAdapter
85
- } from "./chunk-IA2FBTWC.js";
85
+ } from "./chunk-CXDUTWQE.js";
86
86
  import {
87
87
  createBfclAdapter
88
- } from "./chunk-KP5KD6EN.js";
88
+ } from "./chunk-CLIKAXKH.js";
89
89
  import {
90
90
  createCadDesignAdapter
91
91
  } from "./chunk-53UPUNBZ.js";
@@ -141,4 +141,4 @@ export {
141
141
  ADAPTERS,
142
142
  resolveAdapter
143
143
  };
144
- //# sourceMappingURL=chunk-V7AEBY6U.js.map
144
+ //# sourceMappingURL=chunk-EEOC6QPJ.js.map