synth-ai 0.2.14__py3-none-any.whl → 0.4.4__py3-none-any.whl

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.

Potentially problematic release.


This version of synth-ai might be problematic. Click here for more details.

Files changed (1086) hide show
  1. synth_ai/__init__.py +25 -46
  2. synth_ai/__main__.py +30 -3
  3. synth_ai/cli/__init__.py +98 -72
  4. synth_ai/cli/__main__.py +42 -0
  5. synth_ai/cli/_internal/__init__.py +5 -0
  6. synth_ai/cli/_internal/modal_wrapper.py +31 -0
  7. synth_ai/cli/_internal/storage.py +20 -0
  8. synth_ai/cli/_internal/typer_patch.py +47 -0
  9. synth_ai/cli/_internal/validate_task_app.py +29 -0
  10. synth_ai/cli/agents/__init__.py +17 -0
  11. synth_ai/cli/agents/claude.py +77 -0
  12. synth_ai/cli/agents/codex.py +265 -0
  13. synth_ai/cli/agents/opencode.py +253 -0
  14. synth_ai/cli/commands/__init__.py +18 -0
  15. synth_ai/cli/commands/artifacts/__init__.py +13 -0
  16. synth_ai/cli/commands/artifacts/client.py +119 -0
  17. synth_ai/cli/commands/artifacts/config.py +57 -0
  18. synth_ai/cli/commands/artifacts/core.py +24 -0
  19. synth_ai/cli/commands/artifacts/download.py +188 -0
  20. synth_ai/cli/commands/artifacts/export.py +186 -0
  21. synth_ai/cli/commands/artifacts/list.py +156 -0
  22. synth_ai/cli/commands/artifacts/parsing.py +250 -0
  23. synth_ai/cli/commands/artifacts/show.py +336 -0
  24. synth_ai/cli/commands/demo/__init__.py +3 -0
  25. synth_ai/cli/commands/demo/core.py +153 -0
  26. synth_ai/cli/commands/eval/__init__.py +10 -0
  27. synth_ai/cli/commands/eval/config.py +338 -0
  28. synth_ai/cli/commands/eval/core.py +258 -0
  29. synth_ai/cli/commands/eval/runner.py +704 -0
  30. synth_ai/cli/commands/eval/validation.py +60 -0
  31. synth_ai/cli/commands/filter/__init__.py +12 -0
  32. synth_ai/cli/commands/filter/core.py +424 -0
  33. synth_ai/cli/commands/filter/errors.py +55 -0
  34. synth_ai/cli/commands/filter/validation.py +77 -0
  35. synth_ai/cli/commands/help/__init__.py +185 -0
  36. synth_ai/cli/commands/help/core.py +72 -0
  37. synth_ai/cli/commands/scan/__init__.py +19 -0
  38. synth_ai/cli/commands/scan/cloudflare_scanner.py +403 -0
  39. synth_ai/cli/commands/scan/core.py +344 -0
  40. synth_ai/cli/commands/scan/health_checker.py +242 -0
  41. synth_ai/cli/commands/scan/local_scanner.py +278 -0
  42. synth_ai/cli/commands/scan/models.py +83 -0
  43. synth_ai/cli/commands/smoke/__init__.py +7 -0
  44. synth_ai/cli/commands/smoke/core.py +1428 -0
  45. synth_ai/cli/commands/status/__init__.py +3 -0
  46. synth_ai/cli/commands/status/client.py +91 -0
  47. synth_ai/cli/commands/status/config.py +12 -0
  48. synth_ai/cli/commands/status/errors.py +11 -0
  49. synth_ai/cli/commands/status/subcommands/__init__.py +3 -0
  50. synth_ai/cli/commands/status/subcommands/config.py +13 -0
  51. synth_ai/cli/commands/status/subcommands/files.py +34 -0
  52. synth_ai/cli/commands/status/subcommands/jobs.py +51 -0
  53. synth_ai/cli/commands/status/subcommands/models.py +35 -0
  54. synth_ai/cli/commands/status/subcommands/runs.py +34 -0
  55. synth_ai/cli/commands/status/subcommands/session.py +77 -0
  56. synth_ai/cli/commands/status/subcommands/summary.py +39 -0
  57. synth_ai/cli/commands/status/subcommands/utils.py +41 -0
  58. synth_ai/cli/commands/status/utils.py +23 -0
  59. synth_ai/cli/commands/train/__init__.py +51 -0
  60. synth_ai/cli/commands/train/core.py +22 -0
  61. synth_ai/cli/commands/train/errors.py +117 -0
  62. synth_ai/cli/commands/train/prompt_learning_validation.py +632 -0
  63. synth_ai/cli/commands/train/validation.py +392 -0
  64. synth_ai/cli/commands/train/verifier_schemas.py +200 -0
  65. synth_ai/cli/commands/train/verifier_validation.py +235 -0
  66. synth_ai/cli/demo_apps/__init__.py +10 -0
  67. synth_ai/cli/demo_apps/core/__init__.py +28 -0
  68. synth_ai/cli/demo_apps/core/cli.py +1735 -0
  69. synth_ai/cli/demo_apps/crafter/crafter_fft_4b.toml +55 -0
  70. synth_ai/cli/demo_apps/crafter/grpo_crafter_task_app.py +186 -0
  71. synth_ai/cli/demo_apps/crafter/rl_from_base_qwen4b.toml +74 -0
  72. synth_ai/cli/demo_apps/demo_registry.py +176 -0
  73. synth_ai/cli/demo_apps/demo_task_apps/core.py +440 -0
  74. synth_ai/cli/demo_apps/demo_task_apps/crafter/__init__.py +1 -0
  75. synth_ai/cli/demo_apps/demo_task_apps/crafter/grpo_crafter_task_app.py +185 -0
  76. synth_ai/cli/demo_apps/demo_task_apps/math/config.toml +73 -0
  77. synth_ai/cli/demo_apps/demo_task_apps/math/modal_task_app.py +738 -0
  78. synth_ai/cli/demo_apps/demo_task_apps/math/task_app_entry.py +39 -0
  79. synth_ai/cli/demo_apps/math/__init__.py +1 -0
  80. synth_ai/cli/demo_apps/math/_common.py +16 -0
  81. synth_ai/cli/demo_apps/math/app.py +38 -0
  82. synth_ai/cli/demo_apps/math/config.toml +75 -0
  83. synth_ai/cli/demo_apps/math/deploy_modal.py +54 -0
  84. synth_ai/cli/demo_apps/math/modal_task_app.py +698 -0
  85. synth_ai/cli/demo_apps/math/task_app_entry.py +53 -0
  86. synth_ai/cli/demo_apps/mipro/main.py +271 -0
  87. synth_ai/cli/demo_apps/mipro/task_app.py +911 -0
  88. synth_ai/cli/demo_apps/mipro/train_cfg.toml +92 -0
  89. synth_ai/cli/demos/__init__.py +12 -0
  90. synth_ai/cli/demos/demo.py +32 -0
  91. synth_ai/cli/demos/rl_demo.py +254 -0
  92. synth_ai/cli/deploy.py +216 -0
  93. synth_ai/cli/infra/__init__.py +14 -0
  94. synth_ai/cli/infra/balance.py +216 -0
  95. synth_ai/cli/infra/mcp.py +35 -0
  96. synth_ai/cli/infra/modal_app.py +36 -0
  97. synth_ai/cli/infra/setup.py +69 -0
  98. synth_ai/cli/infra/status.py +16 -0
  99. synth_ai/cli/infra/turso.py +77 -0
  100. synth_ai/cli/lib/__init__.py +10 -0
  101. synth_ai/cli/lib/agents.py +76 -0
  102. synth_ai/cli/lib/apps/modal_app.py +101 -0
  103. synth_ai/cli/lib/apps/task_app.py +642 -0
  104. synth_ai/cli/lib/bin.py +39 -0
  105. synth_ai/cli/lib/env.py +375 -0
  106. synth_ai/cli/lib/errors.py +85 -0
  107. synth_ai/cli/lib/modal.py +315 -0
  108. synth_ai/cli/lib/plotting.py +126 -0
  109. synth_ai/cli/lib/prompt_args.py +39 -0
  110. synth_ai/cli/lib/prompts.py +284 -0
  111. synth_ai/cli/lib/sqld.py +122 -0
  112. synth_ai/cli/lib/task_app_discovery.py +884 -0
  113. synth_ai/cli/lib/task_app_env.py +295 -0
  114. synth_ai/cli/lib/train_cfgs.py +300 -0
  115. synth_ai/cli/lib/tunnel_records.py +207 -0
  116. synth_ai/cli/local/__init__.py +14 -0
  117. synth_ai/cli/local/experiment_queue/__init__.py +72 -0
  118. synth_ai/cli/local/experiment_queue/api_schemas.py +221 -0
  119. synth_ai/cli/local/experiment_queue/celery_app.py +208 -0
  120. synth_ai/cli/local/experiment_queue/config.py +128 -0
  121. synth_ai/cli/local/experiment_queue/config_utils.py +272 -0
  122. synth_ai/cli/local/experiment_queue/database.py +175 -0
  123. synth_ai/cli/local/experiment_queue/dispatcher.py +119 -0
  124. synth_ai/cli/local/experiment_queue/models.py +231 -0
  125. synth_ai/cli/local/experiment_queue/progress_info.py +160 -0
  126. synth_ai/cli/local/experiment_queue/results.py +373 -0
  127. synth_ai/cli/local/experiment_queue/schemas.py +131 -0
  128. synth_ai/cli/local/experiment_queue/service.py +344 -0
  129. synth_ai/cli/local/experiment_queue/status.py +372 -0
  130. synth_ai/cli/local/experiment_queue/status_tracker.py +360 -0
  131. synth_ai/cli/local/experiment_queue/tasks.py +1984 -0
  132. synth_ai/cli/local/experiment_queue/trace_storage.py +65 -0
  133. synth_ai/cli/local/experiment_queue/validation.py +157 -0
  134. synth_ai/cli/local/session/__init__.py +92 -0
  135. synth_ai/cli/local/session/client.py +383 -0
  136. synth_ai/cli/local/session/constants.py +63 -0
  137. synth_ai/cli/local/session/exceptions.py +105 -0
  138. synth_ai/cli/local/session/manager.py +139 -0
  139. synth_ai/cli/local/session/models.py +89 -0
  140. synth_ai/cli/local/session/query.py +110 -0
  141. synth_ai/cli/root.py +30 -6
  142. synth_ai/cli/task_apps/__init__.py +37 -0
  143. synth_ai/cli/task_apps/commands.py +3145 -0
  144. synth_ai/cli/task_apps/deploy.py +7 -0
  145. synth_ai/cli/task_apps/list.py +26 -0
  146. synth_ai/cli/task_apps/main.py +36 -0
  147. synth_ai/cli/task_apps/modal_serve.py +11 -0
  148. synth_ai/cli/task_apps/serve.py +11 -0
  149. synth_ai/cli/training/__init__.py +8 -0
  150. synth_ai/cli/training/train.py +5 -0
  151. synth_ai/cli/training/train_cfg.py +34 -0
  152. synth_ai/cli/training/watch.py +506 -0
  153. synth_ai/cli/turso.py +34 -55
  154. synth_ai/cli/utils/__init__.py +8 -0
  155. synth_ai/cli/utils/experiments.py +235 -0
  156. synth_ai/cli/utils/queue.py +504 -0
  157. synth_ai/cli/utils/recent.py +133 -0
  158. synth_ai/cli/utils/traces.py +164 -0
  159. synth_ai/contracts/__init__.py +67 -0
  160. synth_ai/core/__init__.py +100 -0
  161. synth_ai/core/_utils/__init__.py +54 -0
  162. synth_ai/core/_utils/base_url.py +10 -0
  163. synth_ai/core/_utils/http.py +10 -0
  164. synth_ai/core/_utils/prompts.py +14 -0
  165. synth_ai/core/_utils/task_app_state.py +12 -0
  166. synth_ai/core/_utils/user_config.py +10 -0
  167. synth_ai/core/apps/common.py +116 -0
  168. synth_ai/core/auth.py +95 -0
  169. synth_ai/core/cfgs.py +240 -0
  170. synth_ai/core/config/__init__.py +16 -0
  171. synth_ai/core/config/base.py +168 -0
  172. synth_ai/core/config/resolver.py +89 -0
  173. synth_ai/core/env.py +231 -0
  174. synth_ai/core/errors.py +125 -0
  175. synth_ai/core/http.py +230 -0
  176. synth_ai/core/integrations/__init__.py +11 -0
  177. synth_ai/core/integrations/cloudflare.py +1886 -0
  178. synth_ai/core/integrations/mcp/__init__.py +6 -0
  179. synth_ai/core/integrations/mcp/__main__.py +8 -0
  180. synth_ai/core/integrations/mcp/claude.py +36 -0
  181. synth_ai/core/integrations/mcp/main.py +254 -0
  182. synth_ai/core/integrations/mcp/setup.py +100 -0
  183. synth_ai/core/integrations/modal.py +277 -0
  184. synth_ai/core/json.py +72 -0
  185. synth_ai/core/log_filter.py +99 -0
  186. synth_ai/core/logging.py +82 -0
  187. synth_ai/core/paths.py +107 -0
  188. synth_ai/core/pricing.py +109 -0
  189. synth_ai/core/process.py +233 -0
  190. synth_ai/core/ssl.py +25 -0
  191. synth_ai/core/storage/__init__.py +71 -0
  192. synth_ai/core/task_app_state.py +318 -0
  193. synth_ai/core/telemetry.py +282 -0
  194. synth_ai/core/tracing_v3/__init__.py +99 -0
  195. synth_ai/core/tracing_v3/abstractions.py +348 -0
  196. synth_ai/core/tracing_v3/config.py +229 -0
  197. synth_ai/core/tracing_v3/constants.py +21 -0
  198. synth_ai/core/tracing_v3/db_config.py +182 -0
  199. synth_ai/core/tracing_v3/decorators.py +401 -0
  200. synth_ai/core/tracing_v3/llm_call_record_helpers.py +437 -0
  201. synth_ai/core/tracing_v3/migration_helper.py +119 -0
  202. synth_ai/core/tracing_v3/session_tracer.py +542 -0
  203. synth_ai/core/tracing_v3/storage/base.py +211 -0
  204. synth_ai/core/tracing_v3/storage/config.py +109 -0
  205. synth_ai/core/tracing_v3/storage/factory.py +39 -0
  206. synth_ai/core/tracing_v3/trace_utils.py +326 -0
  207. synth_ai/core/tracing_v3/turso/daemon.py +278 -0
  208. synth_ai/core/tracing_v3/turso/models.py +470 -0
  209. synth_ai/core/tracing_v3/turso/native_manager.py +1385 -0
  210. synth_ai/core/tracing_v3/utils.py +108 -0
  211. synth_ai/core/urls.py +18 -0
  212. synth_ai/core/user_config.py +137 -0
  213. synth_ai/core/uvicorn.py +222 -0
  214. synth_ai/data/__init__.py +83 -0
  215. synth_ai/data/enums.py +122 -0
  216. synth_ai/data/rewards.py +249 -0
  217. synth_ai/data/traces.py +35 -0
  218. synth_ai/products/__init__.py +6 -0
  219. synth_ai/products/graph_evolve/__init__.py +45 -0
  220. synth_ai/products/graph_evolve/client.py +226 -0
  221. synth_ai/products/graph_evolve/config.py +591 -0
  222. synth_ai/products/graph_evolve/converters/__init__.py +42 -0
  223. synth_ai/products/graph_evolve/converters/openai_sft.py +484 -0
  224. synth_ai/products/graph_evolve/examples/hotpotqa/config.toml +109 -0
  225. synth_ai/products/graph_evolve/run.py +222 -0
  226. synth_ai/products/graph_gepa/__init__.py +23 -0
  227. synth_ai/products/graph_gepa/converters/__init__.py +19 -0
  228. synth_ai/products/graph_gepa/converters/openai_sft.py +29 -0
  229. synth_ai/sdk/__init__.py +129 -0
  230. synth_ai/sdk/api/__init__.py +1 -0
  231. synth_ai/sdk/api/eval/__init__.py +33 -0
  232. synth_ai/sdk/api/eval/job.py +732 -0
  233. synth_ai/sdk/api/models/supported.py +514 -0
  234. synth_ai/sdk/api/research_agent/__init__.py +296 -0
  235. synth_ai/sdk/api/train/__init__.py +85 -0
  236. synth_ai/sdk/api/train/builders.py +1076 -0
  237. synth_ai/sdk/api/train/cli.py +2196 -0
  238. synth_ai/sdk/api/train/config_finder.py +267 -0
  239. synth_ai/sdk/api/train/configs/__init__.py +67 -0
  240. synth_ai/sdk/api/train/configs/prompt_learning.py +1800 -0
  241. synth_ai/sdk/api/train/configs/rl.py +436 -0
  242. synth_ai/sdk/api/train/configs/sft.py +263 -0
  243. synth_ai/sdk/api/train/configs/shared.py +81 -0
  244. synth_ai/sdk/api/train/context_learning.py +312 -0
  245. synth_ai/sdk/api/train/env_resolver.py +418 -0
  246. synth_ai/sdk/api/train/graph_validators.py +216 -0
  247. synth_ai/sdk/api/train/graphgen.py +1102 -0
  248. synth_ai/sdk/api/train/graphgen_models.py +873 -0
  249. synth_ai/sdk/api/train/graphgen_validators.py +109 -0
  250. synth_ai/sdk/api/train/local_api.py +10 -0
  251. synth_ai/sdk/api/train/pollers.py +160 -0
  252. synth_ai/sdk/api/train/progress/__init__.py +97 -0
  253. synth_ai/sdk/api/train/progress/dataclasses.py +569 -0
  254. synth_ai/sdk/api/train/progress/events.py +326 -0
  255. synth_ai/sdk/api/train/progress/results.py +428 -0
  256. synth_ai/sdk/api/train/progress/tracker.py +641 -0
  257. synth_ai/sdk/api/train/prompt_learning.py +800 -0
  258. synth_ai/sdk/api/train/rl.py +478 -0
  259. synth_ai/sdk/api/train/sft.py +398 -0
  260. synth_ai/sdk/api/train/summary.py +522 -0
  261. synth_ai/sdk/api/train/supported_algos.py +147 -0
  262. synth_ai/sdk/api/train/task_app.py +351 -0
  263. synth_ai/sdk/api/train/utils.py +279 -0
  264. synth_ai/sdk/api/train/validators.py +2424 -0
  265. synth_ai/sdk/graphs/__init__.py +15 -0
  266. synth_ai/sdk/graphs/completions.py +776 -0
  267. synth_ai/sdk/graphs/verifier_schemas.py +222 -0
  268. synth_ai/sdk/inference/__init__.py +6 -0
  269. synth_ai/sdk/inference/client.py +128 -0
  270. synth_ai/sdk/jobs/__init__.py +16 -0
  271. synth_ai/sdk/jobs/client.py +371 -0
  272. synth_ai/sdk/learning/__init__.py +99 -0
  273. synth_ai/sdk/learning/client.py +240 -0
  274. synth_ai/sdk/learning/context_learning_client.py +531 -0
  275. synth_ai/sdk/learning/context_learning_types.py +294 -0
  276. synth_ai/sdk/learning/ft_client.py +7 -0
  277. synth_ai/sdk/learning/health.py +49 -0
  278. synth_ai/sdk/learning/jobs.py +202 -0
  279. synth_ai/sdk/learning/prompt_extraction.py +334 -0
  280. synth_ai/sdk/learning/prompt_learning_client.py +455 -0
  281. synth_ai/sdk/learning/prompt_learning_types.py +186 -0
  282. synth_ai/sdk/learning/rl/__init__.py +35 -0
  283. synth_ai/sdk/learning/rl/client.py +268 -0
  284. synth_ai/sdk/learning/rl/contracts.py +23 -0
  285. synth_ai/sdk/learning/rl/env_keys.py +166 -0
  286. synth_ai/sdk/learning/rl/secrets.py +13 -0
  287. synth_ai/sdk/learning/sft/client.py +95 -0
  288. synth_ai/sdk/learning/sft/config.py +270 -0
  289. synth_ai/sdk/learning/sft/data.py +698 -0
  290. synth_ai/sdk/learning/validators.py +52 -0
  291. synth_ai/sdk/localapi/__init__.py +40 -0
  292. synth_ai/sdk/localapi/apps/__init__.py +28 -0
  293. synth_ai/sdk/localapi/client.py +10 -0
  294. synth_ai/sdk/localapi/contracts.py +10 -0
  295. synth_ai/sdk/localapi/helpers.py +519 -0
  296. synth_ai/sdk/localapi/rollouts.py +93 -0
  297. synth_ai/sdk/localapi/server.py +29 -0
  298. synth_ai/sdk/localapi/template.py +49 -0
  299. synth_ai/sdk/streaming/__init__.py +35 -0
  300. synth_ai/sdk/streaming/config.py +94 -0
  301. synth_ai/sdk/streaming/handlers.py +1997 -0
  302. synth_ai/sdk/streaming/streamer.py +708 -0
  303. synth_ai/sdk/streaming/types.py +112 -0
  304. synth_ai/sdk/task/__init__.py +164 -0
  305. synth_ai/sdk/task/apps/__init__.py +169 -0
  306. synth_ai/sdk/task/client.py +175 -0
  307. synth_ai/sdk/task/config.py +256 -0
  308. synth_ai/sdk/task/contracts.py +340 -0
  309. synth_ai/sdk/task/datasets.py +108 -0
  310. synth_ai/sdk/task/in_process.py +1200 -0
  311. synth_ai/sdk/task/in_process_runner.py +314 -0
  312. synth_ai/sdk/task/inference_api.py +299 -0
  313. synth_ai/sdk/task/proxy.py +287 -0
  314. synth_ai/sdk/task/rubrics/__init__.py +54 -0
  315. synth_ai/sdk/task/rubrics/loaders.py +156 -0
  316. synth_ai/sdk/task/rubrics/strict.py +148 -0
  317. synth_ai/sdk/task/rubrics.py +219 -0
  318. synth_ai/sdk/task/server.py +640 -0
  319. synth_ai/sdk/task/trace_correlation_helpers.py +557 -0
  320. synth_ai/sdk/task/tracing_utils.py +95 -0
  321. synth_ai/sdk/task/validators.py +441 -0
  322. synth_ai/sdk/training/__init__.py +93 -0
  323. synth_ai/sdk/tunnels/__init__.py +118 -0
  324. synth_ai/sdk/tunnels/cleanup.py +83 -0
  325. synth_ai/sdk/tunnels/ports.py +120 -0
  326. synth_ai/sdk/tunnels/tunneled_api.py +363 -0
  327. synth_ai/utils/__init__.py +213 -0
  328. synth_ai-0.4.4.dist-info/METADATA +262 -0
  329. synth_ai-0.4.4.dist-info/RECORD +369 -0
  330. synth_ai-0.4.4.dist-info/top_level.txt +1 -0
  331. examples/__init__.py +0 -16
  332. examples/analyze_semantic_words.sh +0 -17
  333. examples/crafter_debug_render.py +0 -186
  334. examples/dev/qwen3_32b_qlora_4xh100.toml +0 -40
  335. examples/multi_step/configs/README_verilog_rl.md +0 -77
  336. examples/multi_step/configs/VERILOG_REWARDS.md +0 -90
  337. examples/multi_step/configs/VERILOG_RL_CHECKLIST.md +0 -183
  338. examples/multi_step/configs/crafter_eval_synth_qwen4b.toml +0 -35
  339. examples/multi_step/configs/crafter_eval_text_only_groq_qwen32b.toml +0 -36
  340. examples/multi_step/configs/crafter_rl_outcome.toml +0 -74
  341. examples/multi_step/configs/crafter_rl_stepwise_hosted_judge.toml +0 -187
  342. examples/multi_step/configs/crafter_rl_stepwise_shaped.toml +0 -83
  343. examples/multi_step/configs/crafter_rl_stepwise_simple.toml +0 -78
  344. examples/multi_step/configs/crafter_synth_backend.md +0 -40
  345. examples/multi_step/configs/verilog_eval_groq_qwen32b.toml +0 -31
  346. examples/multi_step/configs/verilog_eval_synth_qwen8b.toml +0 -33
  347. examples/multi_step/configs/verilog_rl_lora.toml +0 -190
  348. examples/multi_step/crafter_rl_lora.md +0 -70
  349. examples/multi_step/judges/crafter_backend_judge.py +0 -220
  350. examples/multi_step/judges/verilog_backend_judge.py +0 -234
  351. examples/multi_step/readme.md +0 -48
  352. examples/multi_step/sse_metrics_streaming_notes.md +0 -357
  353. examples/multi_step/task_app_config_notes.md +0 -494
  354. examples/multi_step/verilog_rl_lora.md +0 -218
  355. examples/qwen_coder/README.md +0 -102
  356. examples/qwen_coder/_shared.py +0 -113
  357. examples/qwen_coder/configs/coder_lora_30b.toml +0 -61
  358. examples/qwen_coder/configs/coder_lora_4b.toml +0 -57
  359. examples/qwen_coder/configs/coder_lora_small.toml +0 -58
  360. examples/qwen_coder/generate_dataset.py +0 -98
  361. examples/qwen_coder/infer_ft_smoke.py +0 -65
  362. examples/qwen_coder/infer_prod_proxy.py +0 -73
  363. examples/qwen_coder/infer_via_synth.py +0 -87
  364. examples/qwen_coder/scripts/infer_coder.sh +0 -19
  365. examples/qwen_coder/scripts/train_coder_30b.sh +0 -22
  366. examples/qwen_coder/sft_full_17b.py +0 -103
  367. examples/qwen_coder/sft_lora_30b.py +0 -110
  368. examples/qwen_coder/subset_jsonl.py +0 -39
  369. examples/qwen_coder/todos.md +0 -38
  370. examples/qwen_coder/validate_jsonl.py +0 -60
  371. examples/rl/README.md +0 -169
  372. examples/rl/download_dataset.py +0 -80
  373. examples/run_crafter_demo.sh +0 -10
  374. examples/sft/README.md +0 -139
  375. examples/sft/configs/crafter_fft_qwen0p6b.toml +0 -44
  376. examples/sft/configs/crafter_lora_qwen0p6b.toml +0 -45
  377. examples/sft/evaluate.py +0 -119
  378. examples/sft/export_dataset.py +0 -117
  379. examples/sft/generate_traces.py +0 -164
  380. examples/swe/__init__.py +0 -12
  381. examples/swe/task_app/README.md +0 -105
  382. examples/swe/task_app/__init__.py +0 -2
  383. examples/swe/task_app/grpo_swe_mini.py +0 -601
  384. examples/swe/task_app/grpo_swe_mini_task_app.py +0 -136
  385. examples/swe/task_app/hosted/README.md +0 -173
  386. examples/swe/task_app/hosted/__init__.py +0 -5
  387. examples/swe/task_app/hosted/branching.py +0 -143
  388. examples/swe/task_app/hosted/environment_routes.py +0 -1289
  389. examples/swe/task_app/hosted/envs/__init__.py +0 -1
  390. examples/swe/task_app/hosted/envs/crafter/__init__.py +0 -6
  391. examples/swe/task_app/hosted/envs/crafter/app.py +0 -1
  392. examples/swe/task_app/hosted/envs/crafter/environment.py +0 -522
  393. examples/swe/task_app/hosted/envs/crafter/policy.py +0 -478
  394. examples/swe/task_app/hosted/envs/crafter/react_agent.py +0 -108
  395. examples/swe/task_app/hosted/envs/crafter/shared.py +0 -305
  396. examples/swe/task_app/hosted/envs/crafter/tools.py +0 -47
  397. examples/swe/task_app/hosted/envs/mini_swe/__init__.py +0 -8
  398. examples/swe/task_app/hosted/envs/mini_swe/environment.py +0 -1164
  399. examples/swe/task_app/hosted/envs/mini_swe/policy.py +0 -355
  400. examples/swe/task_app/hosted/envs/mini_swe/shared.py +0 -83
  401. examples/swe/task_app/hosted/envs/mini_swe/tools.py +0 -96
  402. examples/swe/task_app/hosted/hosted_app.py +0 -204
  403. examples/swe/task_app/hosted/inference/__init__.py +0 -5
  404. examples/swe/task_app/hosted/inference/openai_client.py +0 -618
  405. examples/swe/task_app/hosted/main.py +0 -100
  406. examples/swe/task_app/hosted/policy_routes.py +0 -1079
  407. examples/swe/task_app/hosted/registry.py +0 -195
  408. examples/swe/task_app/hosted/rollout.py +0 -1911
  409. examples/swe/task_app/hosted/storage/__init__.py +0 -5
  410. examples/swe/task_app/hosted/storage/volume.py +0 -211
  411. examples/swe/task_app/hosted/test_agents.py +0 -161
  412. examples/swe/task_app/hosted/test_service.py +0 -136
  413. examples/swe/task_app/hosted/utils.py +0 -62
  414. examples/task_apps/IMAGE_ONLY_EVAL_QUICKSTART.md +0 -258
  415. examples/task_apps/TESTING.md +0 -275
  416. examples/task_apps/crafter/CREATE_SFT_DATASET.md +0 -273
  417. examples/task_apps/crafter/EVAL_IMAGE_ONLY_RESULTS.md +0 -152
  418. examples/task_apps/crafter/FILTER_COMMAND_STATUS.md +0 -174
  419. examples/task_apps/crafter/FILTER_COMMAND_SUCCESS.md +0 -268
  420. examples/task_apps/crafter/QUERY_EXAMPLES.md +0 -203
  421. examples/task_apps/crafter/README_IMAGE_ONLY_EVAL.md +0 -316
  422. examples/task_apps/crafter/__init__.py +0 -0
  423. examples/task_apps/crafter/eval_image_only_gpt4o.toml +0 -28
  424. examples/task_apps/crafter/eval_text_only_groq_llama.toml +0 -36
  425. examples/task_apps/crafter/filter_sft_dataset.toml +0 -16
  426. examples/task_apps/crafter/task_app/README.md +0 -42
  427. examples/task_apps/crafter/task_app/__init__.py +0 -5
  428. examples/task_apps/crafter/task_app/grpo_crafter.py +0 -973
  429. examples/task_apps/crafter/task_app/grpo_crafter_task_app.py +0 -146
  430. examples/task_apps/crafter/task_app/synth_envs_hosted/README.md +0 -173
  431. examples/task_apps/crafter/task_app/synth_envs_hosted/__init__.py +0 -5
  432. examples/task_apps/crafter/task_app/synth_envs_hosted/branching.py +0 -143
  433. examples/task_apps/crafter/task_app/synth_envs_hosted/environment_routes.py +0 -1226
  434. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/__init__.py +0 -1
  435. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/crafter/__init__.py +0 -6
  436. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/crafter/app.py +0 -1
  437. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/crafter/environment.py +0 -532
  438. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/crafter/policy.py +0 -547
  439. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/crafter/react_agent.py +0 -123
  440. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/crafter/shared.py +0 -305
  441. examples/task_apps/crafter/task_app/synth_envs_hosted/envs/crafter/tools.py +0 -47
  442. examples/task_apps/crafter/task_app/synth_envs_hosted/hosted_app.py +0 -204
  443. examples/task_apps/crafter/task_app/synth_envs_hosted/inference/__init__.py +0 -5
  444. examples/task_apps/crafter/task_app/synth_envs_hosted/inference/openai_client.py +0 -704
  445. examples/task_apps/crafter/task_app/synth_envs_hosted/main.py +0 -100
  446. examples/task_apps/crafter/task_app/synth_envs_hosted/policy_routes.py +0 -1152
  447. examples/task_apps/crafter/task_app/synth_envs_hosted/registry.py +0 -195
  448. examples/task_apps/crafter/task_app/synth_envs_hosted/rollout.py +0 -2160
  449. examples/task_apps/crafter/task_app/synth_envs_hosted/storage/__init__.py +0 -5
  450. examples/task_apps/crafter/task_app/synth_envs_hosted/storage/volume.py +0 -211
  451. examples/task_apps/crafter/task_app/synth_envs_hosted/test_agents.py +0 -161
  452. examples/task_apps/crafter/task_app/synth_envs_hosted/test_service.py +0 -136
  453. examples/task_apps/crafter/task_app/synth_envs_hosted/utils.py +0 -218
  454. examples/task_apps/dev/pokemon_emerald/__init__.py +0 -2
  455. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/README.md +0 -811
  456. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/agent/__init__.py +0 -120
  457. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/agent/action.py +0 -160
  458. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/agent/memory.py +0 -155
  459. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/agent/perception.py +0 -69
  460. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/agent/planning.py +0 -96
  461. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/agent/simple.py +0 -1502
  462. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/agent/system_prompt.py +0 -4
  463. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/grab_map.py +0 -68
  464. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/manual.py +0 -216
  465. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pokemon_env/__init__.py +0 -35
  466. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pokemon_env/emerald_utils.py +0 -631
  467. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pokemon_env/emulator.py +0 -1544
  468. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pokemon_env/enums.py +0 -1428
  469. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pokemon_env/memory_reader.py +0 -4848
  470. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pokemon_env/types.py +0 -41
  471. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pokemon_env/utils.py +0 -298
  472. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/pyproject.toml +0 -95
  473. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/run.py +0 -204
  474. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/server/__init__.py +0 -0
  475. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/server/app.py +0 -2152
  476. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/server/client.py +0 -429
  477. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/server/frame_server.py +0 -155
  478. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/README.md +0 -78
  479. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/__init__.py +0 -0
  480. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/run_tests.py +0 -122
  481. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_agent_direct.py +0 -76
  482. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_agent_prompts.py +0 -413
  483. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_battle_state_formatting.py +0 -204
  484. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_dialogue_detection.py +0 -133
  485. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_dialogue_detection_comprehensive.py +0 -229
  486. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_direct_agent_emulator.py +0 -300
  487. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_fps_adjustment_pytest.py +0 -205
  488. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_house_to_outside_direct.py +0 -200
  489. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_house_to_outside_transition.py +0 -284
  490. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_map_ground_truth_comparison.py +0 -468
  491. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_memory_map.py +0 -575
  492. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_server_map_validation.py +0 -311
  493. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/tests/test_torchic_state.py +0 -259
  494. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/__init__.py +0 -0
  495. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/anticheat.py +0 -372
  496. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/checkpoint.py +0 -296
  497. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/error_handler.py +0 -275
  498. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/get_local_ip.py +0 -22
  499. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/helpers.py +0 -44
  500. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/llm_logger.py +0 -514
  501. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/map_formatter.py +0 -415
  502. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/map_stitcher.py +0 -1763
  503. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/map_stitcher_singleton.py +0 -33
  504. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/map_trimmer.py +0 -106
  505. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/map_visualizer.py +0 -334
  506. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/ocr_dialogue.py +0 -1020
  507. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/recording.py +0 -188
  508. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/state_formatter.py +0 -1481
  509. examples/task_apps/dev/pokemon_emerald/external/pokeagent-speedrun/utils/vlm.py +0 -862
  510. examples/task_apps/dev/pokemon_emerald/modal_app.py +0 -114
  511. examples/task_apps/dev/pokemon_emerald/task_app/README.md +0 -81
  512. examples/task_apps/dev/pokemon_emerald/task_app/__init__.py +0 -6
  513. examples/task_apps/dev/pokemon_emerald/task_app/pokemon_emerald.py +0 -685
  514. examples/task_apps/enron/__init__.py +0 -1
  515. examples/task_apps/enron/eval_groq_qwen32.toml +0 -16
  516. examples/task_apps/enron/filter_sft.toml +0 -5
  517. examples/task_apps/enron/task_app/README.md +0 -14
  518. examples/task_apps/enron/task_app/__init__.py +0 -1
  519. examples/task_apps/enron/task_app/grpo_enron.py +0 -906
  520. examples/task_apps/enron/task_app/grpo_enron_task_app.py +0 -146
  521. examples/task_apps/enron/tests/__init__.py +0 -4
  522. examples/task_apps/enron/tests/conftest.py +0 -115
  523. examples/task_apps/enron/tests/integration/__init__.py +0 -4
  524. examples/task_apps/enron/tests/integration/test_enron_eval.py +0 -179
  525. examples/task_apps/enron/tests/integration/test_enron_rollout.py +0 -135
  526. examples/task_apps/enron/tests/unit/__init__.py +0 -4
  527. examples/task_apps/enron/tests/unit/test_enron_environment.py +0 -126
  528. examples/task_apps/math/README.md +0 -22
  529. examples/task_apps/math/__init__.py +0 -0
  530. examples/task_apps/math/math_single_step.py +0 -1000
  531. examples/task_apps/math/math_task_app.py +0 -115
  532. examples/task_apps/pokemon_battle/__init__.py +0 -2
  533. examples/task_apps/pokemon_battle/modal_app.py +0 -104
  534. examples/task_apps/pokemon_battle/task_app/README.md +0 -68
  535. examples/task_apps/pokemon_battle/task_app/__init__.py +0 -6
  536. examples/task_apps/pokemon_battle/task_app/pokemon_showdown.py +0 -932
  537. examples/task_apps/pokemon_red/EVAL_IMAGE_ONLY_COMPLETE.md +0 -283
  538. examples/task_apps/pokemon_red/EVAL_IMAGE_ONLY_STATUS.md +0 -155
  539. examples/task_apps/pokemon_red/README.md +0 -357
  540. examples/task_apps/pokemon_red/README_IMAGE_ONLY_EVAL.md +0 -415
  541. examples/task_apps/pokemon_red/__init__.py +0 -3
  542. examples/task_apps/pokemon_red/eval_image_only_gpt4o.toml +0 -29
  543. examples/task_apps/pokemon_red/eval_pokemon_red_policy.py +0 -225
  544. examples/task_apps/pokemon_red/pallet_town_rl_config.toml +0 -75
  545. examples/task_apps/pokemon_red/task_app.py +0 -799
  546. examples/task_apps/pokemon_red/test_pallet_town_rewards.py +0 -193
  547. examples/task_apps/sokoban/README.md +0 -307
  548. examples/task_apps/sokoban/__init__.py +0 -3
  549. examples/task_apps/sokoban/eval_groq_qwen32.toml +0 -16
  550. examples/task_apps/sokoban/eval_openai_gpt5.toml +0 -16
  551. examples/task_apps/sokoban/filter_sft.toml +0 -5
  552. examples/task_apps/sokoban/task_app.py +0 -1058
  553. examples/task_apps/sokoban/tests/__init__.py +0 -4
  554. examples/task_apps/sokoban/tests/conftest.py +0 -113
  555. examples/task_apps/sokoban/tests/integration/__init__.py +0 -4
  556. examples/task_apps/sokoban/tests/integration/test_sokoban_eval.py +0 -57
  557. examples/task_apps/sokoban/tests/integration/test_sokoban_rollout.py +0 -198
  558. examples/task_apps/sokoban/tests/unit/__init__.py +0 -4
  559. examples/task_apps/sokoban/tests/unit/test_sokoban_environment.py +0 -114
  560. examples/task_apps/verilog/__init__.py +0 -1
  561. examples/task_apps/verilog/eval_groq_qwen32b.toml +0 -24
  562. examples/task_apps/verilog/filter_sft.toml +0 -5
  563. examples/task_apps/verilog/task_app/README.md +0 -12
  564. examples/task_apps/verilog/task_app/__init__.py +0 -1
  565. examples/task_apps/verilog/task_app/grpo_verilog.py +0 -1166
  566. examples/task_apps/verilog/task_app/grpo_verilog_task_app.py +0 -145
  567. examples/task_apps/verilog/tests/__init__.py +0 -4
  568. examples/task_apps/verilog/tests/conftest.py +0 -115
  569. examples/task_apps/verilog/tests/integration/__init__.py +0 -4
  570. examples/task_apps/verilog/tests/integration/test_verilog_eval.py +0 -181
  571. examples/task_apps/verilog/tests/integration/test_verilog_rollout.py +0 -55
  572. examples/task_apps/verilog/tests/unit/__init__.py +0 -4
  573. examples/task_apps/verilog/tests/unit/test_verilog_scoring.py +0 -118
  574. examples/vlm/PROPOSAL.md +0 -53
  575. examples/vlm/README.md +0 -68
  576. examples/vlm/configs/crafter_vlm_gpt4o.toml +0 -44
  577. examples/vlm/crafter_image_only_agent.py +0 -207
  578. examples/vlm/crafter_openai_vlm_agent.py +0 -277
  579. examples/vlm/filter_image_rows.py +0 -63
  580. examples/vlm/run_crafter_vlm_benchmark.py +0 -316
  581. examples/warming_up_to_rl/analyze_trace_db.py +0 -422
  582. examples/warming_up_to_rl/configs/crafter_fft.toml +0 -48
  583. examples/warming_up_to_rl/configs/crafter_fft_4b.toml +0 -54
  584. examples/warming_up_to_rl/configs/eval_fft_qwen4b.toml +0 -20
  585. examples/warming_up_to_rl/configs/eval_groq_qwen32b.toml +0 -13
  586. examples/warming_up_to_rl/configs/eval_modal_qwen4b.toml +0 -23
  587. examples/warming_up_to_rl/configs/eval_stepwise_complex.toml +0 -35
  588. examples/warming_up_to_rl/configs/eval_stepwise_consistent.toml +0 -26
  589. examples/warming_up_to_rl/configs/eval_stepwise_per_achievement.toml +0 -36
  590. examples/warming_up_to_rl/configs/eval_stepwise_simple.toml +0 -32
  591. examples/warming_up_to_rl/configs/rl_from_base_qwen4b.toml +0 -83
  592. examples/warming_up_to_rl/configs/rl_from_ft.toml +0 -56
  593. examples/warming_up_to_rl/export_trace_sft.py +0 -723
  594. examples/warming_up_to_rl/groq_test.py +0 -97
  595. examples/warming_up_to_rl/manage_secrets.py +0 -131
  596. examples/warming_up_to_rl/old/event_rewards.md +0 -234
  597. examples/warming_up_to_rl/old/notes.md +0 -73
  598. examples/warming_up_to_rl/readme.md +0 -179
  599. examples/warming_up_to_rl/run_eval.py +0 -736
  600. examples/warming_up_to_rl/run_fft_and_save.py +0 -380
  601. examples/warming_up_to_rl/run_local_rollout.py +0 -239
  602. examples/warming_up_to_rl/run_local_rollout_modal.py +0 -248
  603. examples/warming_up_to_rl/run_local_rollout_parallel.py +0 -405
  604. examples/warming_up_to_rl/run_local_rollout_traced.py +0 -477
  605. examples/warming_up_to_rl/run_rl_and_save.py +0 -124
  606. examples/warming_up_to_rl/run_rollout_remote.py +0 -156
  607. examples/workflows/__init__.py +0 -0
  608. examples/workflows/math_rl/__init__.py +0 -0
  609. examples/workflows/math_rl/configs/eval_base_qwen.toml +0 -15
  610. examples/workflows/math_rl/configs/eval_rl_qwen.toml +0 -11
  611. examples/workflows/math_rl/configs/rl_from_base_qwen.toml +0 -35
  612. examples/workflows/math_rl/configs/rl_from_base_qwen17.toml +0 -74
  613. examples/workflows/math_rl/configs/rl_from_ft_qwen.toml +0 -35
  614. examples/workflows/math_rl/download_dataset.py +0 -80
  615. examples/workflows/math_rl/run_eval.py +0 -436
  616. examples/workflows/math_rl/run_rl_and_save.py +0 -111
  617. synth_ai/api/models/supported.py +0 -377
  618. synth_ai/api/train/__init__.py +0 -5
  619. synth_ai/api/train/builders.py +0 -351
  620. synth_ai/api/train/cli.py +0 -635
  621. synth_ai/api/train/config_finder.py +0 -228
  622. synth_ai/api/train/configs/__init__.py +0 -44
  623. synth_ai/api/train/configs/rl.py +0 -134
  624. synth_ai/api/train/configs/sft.py +0 -95
  625. synth_ai/api/train/configs/shared.py +0 -24
  626. synth_ai/api/train/env_resolver.py +0 -349
  627. synth_ai/api/train/pollers.py +0 -75
  628. synth_ai/api/train/supported_algos.py +0 -147
  629. synth_ai/api/train/task_app.py +0 -195
  630. synth_ai/api/train/utils.py +0 -225
  631. synth_ai/cli/_modal_wrapper.py +0 -29
  632. synth_ai/cli/_storage.py +0 -20
  633. synth_ai/cli/_typer_patch.py +0 -49
  634. synth_ai/cli/_validate_task_app.py +0 -11
  635. synth_ai/cli/balance.py +0 -216
  636. synth_ai/cli/calc.py +0 -84
  637. synth_ai/cli/demo.py +0 -165
  638. synth_ai/cli/legacy_root_backup.py +0 -468
  639. synth_ai/cli/man.py +0 -106
  640. synth_ai/cli/recent.py +0 -132
  641. synth_ai/cli/rl_demo.py +0 -254
  642. synth_ai/cli/status.py +0 -134
  643. synth_ai/cli/task_apps.py +0 -4523
  644. synth_ai/cli/traces.py +0 -164
  645. synth_ai/cli/tui.py +0 -57
  646. synth_ai/cli/watch.py +0 -506
  647. synth_ai/compound/cais.py +0 -0
  648. synth_ai/config/base_url.py +0 -107
  649. synth_ai/core/experiment.py +0 -13
  650. synth_ai/core/system.py +0 -15
  651. synth_ai/demo_registry.py +0 -295
  652. synth_ai/demos/core/__init__.py +0 -1
  653. synth_ai/demos/core/cli.py +0 -1718
  654. synth_ai/demos/demo_task_apps/core.py +0 -440
  655. synth_ai/demos/demo_task_apps/crafter/grpo_crafter_task_app.py +0 -184
  656. synth_ai/demos/demo_task_apps/math/config.toml +0 -74
  657. synth_ai/demos/demo_task_apps/math/deploy_task_app.sh +0 -22
  658. synth_ai/demos/demo_task_apps/math/modal_task_app.py +0 -739
  659. synth_ai/demos/demo_task_apps/math/task_app_entry.py +0 -37
  660. synth_ai/environments/__init__.py +0 -31
  661. synth_ai/environments/environment/__init__.py +0 -1
  662. synth_ai/environments/environment/artifacts/__init__.py +0 -1
  663. synth_ai/environments/environment/artifacts/base.py +0 -52
  664. synth_ai/environments/environment/core.py +0 -67
  665. synth_ai/environments/environment/db/__init__.py +0 -1
  666. synth_ai/environments/environment/db/sqlite.py +0 -45
  667. synth_ai/environments/environment/registry.py +0 -233
  668. synth_ai/environments/environment/resources/sqlite.py +0 -45
  669. synth_ai/environments/environment/results.py +0 -1
  670. synth_ai/environments/environment/rewards/__init__.py +0 -1
  671. synth_ai/environments/environment/rewards/core.py +0 -29
  672. synth_ai/environments/environment/shared_engine.py +0 -26
  673. synth_ai/environments/environment/tools/__init__.py +0 -200
  674. synth_ai/environments/examples/__init__.py +0 -1
  675. synth_ai/environments/examples/bandit/__init__.py +0 -33
  676. synth_ai/environments/examples/bandit/engine.py +0 -302
  677. synth_ai/environments/examples/bandit/environment.py +0 -194
  678. synth_ai/environments/examples/bandit/taskset.py +0 -200
  679. synth_ai/environments/examples/crafter_classic/__init__.py +0 -8
  680. synth_ai/environments/examples/crafter_classic/agent_demos/analyze_semantic_words_markdown.py +0 -250
  681. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_comprehensive_evaluation.py +0 -59
  682. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_evaluation_browser.py +0 -152
  683. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_evaluation_config.toml +0 -24
  684. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_evaluation_framework.py +0 -1194
  685. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/crafter_synth_config.toml +0 -56
  686. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/filter_config_modal.toml +0 -32
  687. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/filter_traces_sft_turso.py +0 -738
  688. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/kick_off_ft_modal.py +0 -384
  689. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/analyze_action_results.py +0 -53
  690. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/analyze_agent_actions.py +0 -178
  691. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/analyze_latest_run.py +0 -222
  692. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/analyze_lm_traces.py +0 -183
  693. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/analyze_no_rewards.py +0 -210
  694. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/analyze_trace_issue.py +0 -206
  695. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/check_db_schema.py +0 -49
  696. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/check_latest_results.py +0 -64
  697. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/debug_agent_responses.py +0 -88
  698. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_modal_ft/old/quick_trace_check.py +0 -77
  699. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/compare_experiments.py +0 -324
  700. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/filter_traces_sft_turso.py +0 -580
  701. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/kick_off_ft_oai.py +0 -362
  702. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/multi_model_config.toml +0 -49
  703. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/analyze_enhanced_hooks.py +0 -332
  704. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/analyze_hook_events.py +0 -97
  705. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/analyze_hook_results.py +0 -217
  706. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/check_hook_storage.py +0 -87
  707. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/check_seeds.py +0 -88
  708. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/compare_seed_performance.py +0 -195
  709. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/custom_eval_pipelines.py +0 -400
  710. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/plot_hook_frequency.py +0 -195
  711. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/old/seed_analysis_summary.py +0 -56
  712. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_openai_ft/run_rollouts_for_models_and_compare_v3.py +0 -858
  713. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_quick_evaluation.py +0 -52
  714. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_react_agent.py +0 -874
  715. synth_ai/environments/examples/crafter_classic/agent_demos/crafter_trace_evaluation.py +0 -1412
  716. synth_ai/environments/examples/crafter_classic/agent_demos/example_v3_usage.py +0 -216
  717. synth_ai/environments/examples/crafter_classic/agent_demos/old/compare_traces.py +0 -296
  718. synth_ai/environments/examples/crafter_classic/agent_demos/old/crafter_comprehensive_evaluation.py +0 -58
  719. synth_ai/environments/examples/crafter_classic/agent_demos/old/crafter_env_serialization.py +0 -464
  720. synth_ai/environments/examples/crafter_classic/agent_demos/old/crafter_evaluation_browser.py +0 -152
  721. synth_ai/environments/examples/crafter_classic/agent_demos/old/crafter_quick_evaluation.py +0 -51
  722. synth_ai/environments/examples/crafter_classic/agent_demos/old/crafter_trace_evaluation.py +0 -1412
  723. synth_ai/environments/examples/crafter_classic/agent_demos/old/debug_player_loss.py +0 -112
  724. synth_ai/environments/examples/crafter_classic/agent_demos/old/diagnose_service.py +0 -203
  725. synth_ai/environments/examples/crafter_classic/agent_demos/old/diagnose_slowness.py +0 -305
  726. synth_ai/environments/examples/crafter_classic/agent_demos/old/eval_by_difficulty.py +0 -126
  727. synth_ai/environments/examples/crafter_classic/agent_demos/old/eval_example.py +0 -94
  728. synth_ai/environments/examples/crafter_classic/agent_demos/old/explore_saved_states.py +0 -142
  729. synth_ai/environments/examples/crafter_classic/agent_demos/old/filter_traces_sft.py +0 -26
  730. synth_ai/environments/examples/crafter_classic/agent_demos/old/filter_traces_sft_OLD.py +0 -984
  731. synth_ai/environments/examples/crafter_classic/agent_demos/old/generate_ft_data_gemini.py +0 -724
  732. synth_ai/environments/examples/crafter_classic/agent_demos/old/generate_ft_data_modal.py +0 -386
  733. synth_ai/environments/examples/crafter_classic/agent_demos/old/generate_ft_metadata.py +0 -205
  734. synth_ai/environments/examples/crafter_classic/agent_demos/old/kick_off_ft_gemini.py +0 -150
  735. synth_ai/environments/examples/crafter_classic/agent_demos/old/kick_off_ft_modal.py +0 -283
  736. synth_ai/environments/examples/crafter_classic/agent_demos/old/prepare_vertex_ft.py +0 -280
  737. synth_ai/environments/examples/crafter_classic/agent_demos/old/profile_env_slowness.py +0 -456
  738. synth_ai/environments/examples/crafter_classic/agent_demos/old/replicate_issue.py +0 -166
  739. synth_ai/environments/examples/crafter_classic/agent_demos/old/run_and_eval.py +0 -102
  740. synth_ai/environments/examples/crafter_classic/agent_demos/old/run_comparison.py +0 -128
  741. synth_ai/environments/examples/crafter_classic/agent_demos/old/run_qwen_rollouts.py +0 -655
  742. synth_ai/environments/examples/crafter_classic/agent_demos/old/trace_eval_OLD.py +0 -202
  743. synth_ai/environments/examples/crafter_classic/agent_demos/old/validate_openai_format.py +0 -166
  744. synth_ai/environments/examples/crafter_classic/config_logging.py +0 -111
  745. synth_ai/environments/examples/crafter_classic/debug_translation.py +0 -0
  746. synth_ai/environments/examples/crafter_classic/engine.py +0 -579
  747. synth_ai/environments/examples/crafter_classic/engine_deterministic_patch.py +0 -64
  748. synth_ai/environments/examples/crafter_classic/engine_helpers/action_map.py +0 -6
  749. synth_ai/environments/examples/crafter_classic/engine_helpers/serialization.py +0 -75
  750. synth_ai/environments/examples/crafter_classic/engine_serialization_patch_v3.py +0 -267
  751. synth_ai/environments/examples/crafter_classic/environment.py +0 -495
  752. synth_ai/environments/examples/crafter_classic/taskset.py +0 -233
  753. synth_ai/environments/examples/crafter_classic/trace_hooks_v3.py +0 -228
  754. synth_ai/environments/examples/crafter_classic/world_config_patch_simple.py +0 -299
  755. synth_ai/environments/examples/crafter_custom/__init__.py +0 -4
  756. synth_ai/environments/examples/crafter_custom/agent_demos/__init__.py +0 -1
  757. synth_ai/environments/examples/crafter_custom/agent_demos/trace_eval.py +0 -202
  758. synth_ai/environments/examples/crafter_custom/crafter/__init__.py +0 -7
  759. synth_ai/environments/examples/crafter_custom/crafter/config.py +0 -182
  760. synth_ai/environments/examples/crafter_custom/crafter/constants.py +0 -8
  761. synth_ai/environments/examples/crafter_custom/crafter/engine.py +0 -269
  762. synth_ai/environments/examples/crafter_custom/crafter/env.py +0 -262
  763. synth_ai/environments/examples/crafter_custom/crafter/objects.py +0 -417
  764. synth_ai/environments/examples/crafter_custom/crafter/recorder.py +0 -187
  765. synth_ai/environments/examples/crafter_custom/crafter/worldgen.py +0 -118
  766. synth_ai/environments/examples/crafter_custom/dataset_builder.py +0 -373
  767. synth_ai/environments/examples/crafter_custom/environment.py +0 -312
  768. synth_ai/environments/examples/crafter_custom/old/analyze_diamond_issue.py +0 -159
  769. synth_ai/environments/examples/crafter_custom/old/analyze_diamond_spawning.py +0 -158
  770. synth_ai/environments/examples/crafter_custom/old/compare_worlds.py +0 -71
  771. synth_ai/environments/examples/crafter_custom/old/dataset_stats.py +0 -105
  772. synth_ai/environments/examples/crafter_custom/old/diamond_spawning_summary.py +0 -119
  773. synth_ai/environments/examples/crafter_custom/old/example_dataset_usage.py +0 -52
  774. synth_ai/environments/examples/crafter_custom/run_dataset.py +0 -305
  775. synth_ai/environments/examples/enron/art_helpers/email_search_tools.py +0 -156
  776. synth_ai/environments/examples/enron/art_helpers/local_email_db.py +0 -281
  777. synth_ai/environments/examples/enron/art_helpers/types_enron.py +0 -25
  778. synth_ai/environments/examples/enron/engine.py +0 -300
  779. synth_ai/environments/examples/enron/environment.py +0 -234
  780. synth_ai/environments/examples/enron/taskset.py +0 -112
  781. synth_ai/environments/examples/enron/units/keyword_stats.py +0 -112
  782. synth_ai/environments/examples/minigrid/__init__.py +0 -48
  783. synth_ai/environments/examples/minigrid/agent_demos/minigrid_evaluation_framework.py +0 -1188
  784. synth_ai/environments/examples/minigrid/agent_demos/minigrid_quick_evaluation.py +0 -48
  785. synth_ai/environments/examples/minigrid/agent_demos/minigrid_react_agent.py +0 -562
  786. synth_ai/environments/examples/minigrid/agent_demos/minigrid_trace_evaluation.py +0 -221
  787. synth_ai/environments/examples/minigrid/engine.py +0 -589
  788. synth_ai/environments/examples/minigrid/environment.py +0 -274
  789. synth_ai/environments/examples/minigrid/environment_mapping.py +0 -242
  790. synth_ai/environments/examples/minigrid/puzzle_loader.py +0 -417
  791. synth_ai/environments/examples/minigrid/taskset.py +0 -583
  792. synth_ai/environments/examples/nethack/__init__.py +0 -7
  793. synth_ai/environments/examples/nethack/achievements.py +0 -337
  794. synth_ai/environments/examples/nethack/agent_demos/nethack_evaluation_framework.py +0 -981
  795. synth_ai/environments/examples/nethack/agent_demos/nethack_quick_evaluation.py +0 -74
  796. synth_ai/environments/examples/nethack/agent_demos/nethack_react_agent.py +0 -831
  797. synth_ai/environments/examples/nethack/engine.py +0 -739
  798. synth_ai/environments/examples/nethack/environment.py +0 -256
  799. synth_ai/environments/examples/nethack/helpers/__init__.py +0 -41
  800. synth_ai/environments/examples/nethack/helpers/action_mapping.py +0 -301
  801. synth_ai/environments/examples/nethack/helpers/nle_wrapper.py +0 -402
  802. synth_ai/environments/examples/nethack/helpers/observation_utils.py +0 -433
  803. synth_ai/environments/examples/nethack/helpers/recording_wrapper.py +0 -200
  804. synth_ai/environments/examples/nethack/helpers/trajectory_recorder.py +0 -269
  805. synth_ai/environments/examples/nethack/helpers/visualization/replay_viewer.py +0 -308
  806. synth_ai/environments/examples/nethack/helpers/visualization/visualizer.py +0 -431
  807. synth_ai/environments/examples/nethack/taskset.py +0 -323
  808. synth_ai/environments/examples/red/__init__.py +0 -7
  809. synth_ai/environments/examples/red/agent_demos/__init__.py +0 -1
  810. synth_ai/environments/examples/red/config_logging.py +0 -110
  811. synth_ai/environments/examples/red/engine.py +0 -721
  812. synth_ai/environments/examples/red/engine_helpers/__init__.py +0 -1
  813. synth_ai/environments/examples/red/engine_helpers/memory_map.py +0 -35
  814. synth_ai/environments/examples/red/engine_helpers/reward_components.py +0 -276
  815. synth_ai/environments/examples/red/engine_helpers/reward_library/__init__.py +0 -142
  816. synth_ai/environments/examples/red/engine_helpers/reward_library/adaptive_rewards.py +0 -57
  817. synth_ai/environments/examples/red/engine_helpers/reward_library/battle_rewards.py +0 -284
  818. synth_ai/environments/examples/red/engine_helpers/reward_library/composite_rewards.py +0 -150
  819. synth_ai/environments/examples/red/engine_helpers/reward_library/economy_rewards.py +0 -138
  820. synth_ai/environments/examples/red/engine_helpers/reward_library/efficiency_rewards.py +0 -57
  821. synth_ai/environments/examples/red/engine_helpers/reward_library/exploration_rewards.py +0 -331
  822. synth_ai/environments/examples/red/engine_helpers/reward_library/novelty_rewards.py +0 -121
  823. synth_ai/environments/examples/red/engine_helpers/reward_library/pallet_town_progression.py +0 -477
  824. synth_ai/environments/examples/red/engine_helpers/reward_library/pallet_town_rewards.py +0 -559
  825. synth_ai/environments/examples/red/engine_helpers/reward_library/pokemon_rewards.py +0 -313
  826. synth_ai/environments/examples/red/engine_helpers/reward_library/social_rewards.py +0 -148
  827. synth_ai/environments/examples/red/engine_helpers/reward_library/story_rewards.py +0 -247
  828. synth_ai/environments/examples/red/engine_helpers/screen_analysis.py +0 -368
  829. synth_ai/environments/examples/red/engine_helpers/state_extraction.py +0 -172
  830. synth_ai/environments/examples/red/environment.py +0 -298
  831. synth_ai/environments/examples/red/taskset.py +0 -79
  832. synth_ai/environments/examples/red/units/__init__.py +0 -1
  833. synth_ai/environments/examples/sokoban/__init__.py +0 -1
  834. synth_ai/environments/examples/sokoban/agent_demos/sokoban_full_eval.py +0 -899
  835. synth_ai/environments/examples/sokoban/engine.py +0 -678
  836. synth_ai/environments/examples/sokoban/engine_helpers/__init__.py +0 -1
  837. synth_ai/environments/examples/sokoban/engine_helpers/room_utils.py +0 -657
  838. synth_ai/environments/examples/sokoban/engine_helpers/vendored/__init__.py +0 -18
  839. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/__init__.py +0 -3
  840. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/boxoban_env.py +0 -131
  841. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/render_utils.py +0 -370
  842. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/room_utils.py +0 -332
  843. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/sokoban_env.py +0 -306
  844. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/sokoban_env_fixed_targets.py +0 -67
  845. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/sokoban_env_pull.py +0 -115
  846. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/sokoban_env_two_player.py +0 -123
  847. synth_ai/environments/examples/sokoban/engine_helpers/vendored/envs/sokoban_env_variations.py +0 -394
  848. synth_ai/environments/examples/sokoban/environment.py +0 -229
  849. synth_ai/environments/examples/sokoban/generate_verified_puzzles.py +0 -440
  850. synth_ai/environments/examples/sokoban/puzzle_loader.py +0 -312
  851. synth_ai/environments/examples/sokoban/taskset.py +0 -544
  852. synth_ai/environments/examples/tictactoe/__init__.py +0 -1
  853. synth_ai/environments/examples/tictactoe/engine.py +0 -368
  854. synth_ai/environments/examples/tictactoe/environment.py +0 -240
  855. synth_ai/environments/examples/tictactoe/taskset.py +0 -215
  856. synth_ai/environments/examples/verilog/__init__.py +0 -10
  857. synth_ai/environments/examples/verilog/engine.py +0 -421
  858. synth_ai/environments/examples/verilog/environment.py +0 -350
  859. synth_ai/environments/examples/verilog/taskset.py +0 -420
  860. synth_ai/environments/examples/wordle/__init__.py +0 -29
  861. synth_ai/environments/examples/wordle/engine.py +0 -398
  862. synth_ai/environments/examples/wordle/environment.py +0 -159
  863. synth_ai/environments/examples/wordle/helpers/generate_instances_wordfreq.py +0 -75
  864. synth_ai/environments/examples/wordle/taskset.py +0 -230
  865. synth_ai/environments/reproducibility/core.py +0 -42
  866. synth_ai/environments/reproducibility/helpers.py +0 -0
  867. synth_ai/environments/reproducibility/tree.py +0 -363
  868. synth_ai/environments/service/app.py +0 -97
  869. synth_ai/environments/service/core_routes.py +0 -1021
  870. synth_ai/environments/service/external_registry.py +0 -56
  871. synth_ai/environments/service/registry.py +0 -9
  872. synth_ai/environments/stateful/__init__.py +0 -1
  873. synth_ai/environments/stateful/core.py +0 -163
  874. synth_ai/environments/stateful/engine.py +0 -21
  875. synth_ai/environments/stateful/state.py +0 -7
  876. synth_ai/environments/tasks/api.py +0 -19
  877. synth_ai/environments/tasks/core.py +0 -81
  878. synth_ai/environments/tasks/filters.py +0 -40
  879. synth_ai/environments/tasks/utils.py +0 -90
  880. synth_ai/environments/v0_observability/history.py +0 -3
  881. synth_ai/environments/v0_observability/log.py +0 -2
  882. synth_ai/evals/__init__.py +0 -15
  883. synth_ai/evals/base.py +0 -13
  884. synth_ai/evals/client.py +0 -82
  885. synth_ai/evals/types.py +0 -42
  886. synth_ai/handshake.py +0 -109
  887. synth_ai/http.py +0 -26
  888. synth_ai/http_client.py +0 -136
  889. synth_ai/inference/__init__.py +0 -5
  890. synth_ai/inference/client.py +0 -34
  891. synth_ai/jobs/client.py +0 -295
  892. synth_ai/judge_schemas.py +0 -127
  893. synth_ai/learning/__init__.py +0 -59
  894. synth_ai/learning/client.py +0 -241
  895. synth_ai/learning/ft_client.py +0 -7
  896. synth_ai/learning/health.py +0 -49
  897. synth_ai/learning/jobs.py +0 -201
  898. synth_ai/learning/rl/__init__.py +0 -39
  899. synth_ai/learning/rl/client.py +0 -267
  900. synth_ai/learning/rl/contracts.py +0 -27
  901. synth_ai/learning/rl/env_keys.py +0 -166
  902. synth_ai/learning/rl/secrets.py +0 -13
  903. synth_ai/learning/sft/client.py +0 -68
  904. synth_ai/learning/sft/config.py +0 -270
  905. synth_ai/learning/sft/data.py +0 -295
  906. synth_ai/learning/validators.py +0 -49
  907. synth_ai/lm/__init__.py +0 -25
  908. synth_ai/task/__init__.py +0 -121
  909. synth_ai/task/apps/__init__.py +0 -129
  910. synth_ai/task/client.py +0 -167
  911. synth_ai/task/config.py +0 -257
  912. synth_ai/task/contracts.py +0 -236
  913. synth_ai/task/datasets.py +0 -108
  914. synth_ai/task/proxy.py +0 -251
  915. synth_ai/task/rubrics/__init__.py +0 -56
  916. synth_ai/task/rubrics/loaders.py +0 -152
  917. synth_ai/task/rubrics/strict.py +0 -149
  918. synth_ai/task/server.py +0 -432
  919. synth_ai/task/trace_correlation_helpers.py +0 -315
  920. synth_ai/task/tracing_utils.py +0 -84
  921. synth_ai/task/validators.py +0 -418
  922. synth_ai/tracing_v3/__init__.py +0 -97
  923. synth_ai/tracing_v3/abstractions.py +0 -302
  924. synth_ai/tracing_v3/config.py +0 -84
  925. synth_ai/tracing_v3/db_config.py +0 -194
  926. synth_ai/tracing_v3/decorators.py +0 -398
  927. synth_ai/tracing_v3/llm_call_record_helpers.py +0 -391
  928. synth_ai/tracing_v3/migration_helper.py +0 -120
  929. synth_ai/tracing_v3/session_tracer.py +0 -540
  930. synth_ai/tracing_v3/storage/base.py +0 -210
  931. synth_ai/tracing_v3/storage/config.py +0 -75
  932. synth_ai/tracing_v3/storage/factory.py +0 -39
  933. synth_ai/tracing_v3/trace_utils.py +0 -317
  934. synth_ai/tracing_v3/turso/daemon.py +0 -151
  935. synth_ai/tracing_v3/turso/models.py +0 -469
  936. synth_ai/tracing_v3/turso/native_manager.py +0 -1209
  937. synth_ai/tracing_v3/utils.py +0 -108
  938. synth_ai/tui/__init__.py +0 -5
  939. synth_ai/tui/__main__.py +0 -13
  940. synth_ai/tui/cli/__init__.py +0 -1
  941. synth_ai/tui/cli/query_experiments.py +0 -164
  942. synth_ai/tui/cli/query_experiments_v3.py +0 -164
  943. synth_ai/tui/dashboard.py +0 -906
  944. synth_ai/v0/api/__init__.py +0 -8
  945. synth_ai/v0/api/models/__init__.py +0 -8
  946. synth_ai/v0/api/models/supported.py +0 -8
  947. synth_ai/v0/config/__init__.py +0 -15
  948. synth_ai/v0/config/base_url.py +0 -12
  949. synth_ai/v0/lm/__init__.py +0 -51
  950. synth_ai/v0/lm/caching/__init__.py +0 -0
  951. synth_ai/v0/lm/caching/constants.py +0 -6
  952. synth_ai/v0/lm/caching/dbs.py +0 -0
  953. synth_ai/v0/lm/caching/ephemeral.py +0 -100
  954. synth_ai/v0/lm/caching/handler.py +0 -137
  955. synth_ai/v0/lm/caching/initialize.py +0 -11
  956. synth_ai/v0/lm/caching/persistent.py +0 -114
  957. synth_ai/v0/lm/config.py +0 -115
  958. synth_ai/v0/lm/constants.py +0 -32
  959. synth_ai/v0/lm/core/__init__.py +0 -8
  960. synth_ai/v0/lm/core/all.py +0 -73
  961. synth_ai/v0/lm/core/exceptions.py +0 -5
  962. synth_ai/v0/lm/core/main.py +0 -331
  963. synth_ai/v0/lm/core/main_v3.py +0 -594
  964. synth_ai/v0/lm/core/synth_models.py +0 -35
  965. synth_ai/v0/lm/core/vendor_clients.py +0 -190
  966. synth_ai/v0/lm/cost/__init__.py +0 -0
  967. synth_ai/v0/lm/cost/monitor.py +0 -1
  968. synth_ai/v0/lm/cost/statefulness.py +0 -1
  969. synth_ai/v0/lm/injection.py +0 -80
  970. synth_ai/v0/lm/overrides.py +0 -206
  971. synth_ai/v0/lm/provider_support/__init__.py +0 -8
  972. synth_ai/v0/lm/provider_support/anthropic.py +0 -972
  973. synth_ai/v0/lm/provider_support/openai.py +0 -1139
  974. synth_ai/v0/lm/provider_support/suppress_logging.py +0 -31
  975. synth_ai/v0/lm/structured_outputs/__init__.py +0 -0
  976. synth_ai/v0/lm/structured_outputs/handler.py +0 -440
  977. synth_ai/v0/lm/structured_outputs/inject.py +0 -297
  978. synth_ai/v0/lm/structured_outputs/rehabilitate.py +0 -185
  979. synth_ai/v0/lm/tools/__init__.py +0 -3
  980. synth_ai/v0/lm/tools/base.py +0 -172
  981. synth_ai/v0/lm/unified_interface.py +0 -202
  982. synth_ai/v0/lm/vendors/__init__.py +0 -0
  983. synth_ai/v0/lm/vendors/base.py +0 -81
  984. synth_ai/v0/lm/vendors/core/__init__.py +0 -0
  985. synth_ai/v0/lm/vendors/core/anthropic_api.py +0 -387
  986. synth_ai/v0/lm/vendors/core/gemini_api.py +0 -292
  987. synth_ai/v0/lm/vendors/core/mistral_api.py +0 -322
  988. synth_ai/v0/lm/vendors/core/openai_api.py +0 -227
  989. synth_ai/v0/lm/vendors/core/synth_dev_api.py +0 -0
  990. synth_ai/v0/lm/vendors/local/__init__.py +0 -0
  991. synth_ai/v0/lm/vendors/local/ollama.py +0 -0
  992. synth_ai/v0/lm/vendors/openai_standard.py +0 -782
  993. synth_ai/v0/lm/vendors/openai_standard_responses.py +0 -259
  994. synth_ai/v0/lm/vendors/retries.py +0 -22
  995. synth_ai/v0/lm/vendors/supported/__init__.py +0 -0
  996. synth_ai/v0/lm/vendors/supported/custom_endpoint.py +0 -415
  997. synth_ai/v0/lm/vendors/supported/deepseek.py +0 -69
  998. synth_ai/v0/lm/vendors/supported/grok.py +0 -75
  999. synth_ai/v0/lm/vendors/supported/groq.py +0 -16
  1000. synth_ai/v0/lm/vendors/supported/ollama.py +0 -15
  1001. synth_ai/v0/lm/vendors/supported/openrouter.py +0 -74
  1002. synth_ai/v0/lm/vendors/supported/together.py +0 -11
  1003. synth_ai/v0/lm/vendors/synth_client.py +0 -835
  1004. synth_ai/v0/lm/warmup.py +0 -186
  1005. synth_ai/v0/tracing/__init__.py +0 -0
  1006. synth_ai/v0/tracing/abstractions.py +0 -224
  1007. synth_ai/v0/tracing/base_client.py +0 -91
  1008. synth_ai/v0/tracing/client_manager.py +0 -131
  1009. synth_ai/v0/tracing/config.py +0 -142
  1010. synth_ai/v0/tracing/context.py +0 -146
  1011. synth_ai/v0/tracing/decorators.py +0 -682
  1012. synth_ai/v0/tracing/events/__init__.py +0 -0
  1013. synth_ai/v0/tracing/events/manage.py +0 -147
  1014. synth_ai/v0/tracing/events/scope.py +0 -86
  1015. synth_ai/v0/tracing/events/store.py +0 -228
  1016. synth_ai/v0/tracing/immediate_client.py +0 -151
  1017. synth_ai/v0/tracing/local.py +0 -18
  1018. synth_ai/v0/tracing/log_client_base.py +0 -73
  1019. synth_ai/v0/tracing/retry_queue.py +0 -186
  1020. synth_ai/v0/tracing/trackers.py +0 -515
  1021. synth_ai/v0/tracing/upload.py +0 -409
  1022. synth_ai/v0/tracing/utils.py +0 -9
  1023. synth_ai/v0/tracing_v1/__init__.py +0 -16
  1024. synth_ai/v0/tracing_v1/abstractions.py +0 -224
  1025. synth_ai/v0/tracing_v1/base_client.py +0 -91
  1026. synth_ai/v0/tracing_v1/client_manager.py +0 -131
  1027. synth_ai/v0/tracing_v1/config.py +0 -142
  1028. synth_ai/v0/tracing_v1/context.py +0 -146
  1029. synth_ai/v0/tracing_v1/decorators.py +0 -703
  1030. synth_ai/v0/tracing_v1/events/__init__.py +0 -0
  1031. synth_ai/v0/tracing_v1/events/manage.py +0 -147
  1032. synth_ai/v0/tracing_v1/events/scope.py +0 -86
  1033. synth_ai/v0/tracing_v1/events/store.py +0 -228
  1034. synth_ai/v0/tracing_v1/immediate_client.py +0 -151
  1035. synth_ai/v0/tracing_v1/local.py +0 -18
  1036. synth_ai/v0/tracing_v1/log_client_base.py +0 -73
  1037. synth_ai/v0/tracing_v1/retry_queue.py +0 -186
  1038. synth_ai/v0/tracing_v1/trackers.py +0 -515
  1039. synth_ai/v0/tracing_v1/upload.py +0 -527
  1040. synth_ai/v0/tracing_v1/utils.py +0 -9
  1041. synth_ai/v0/tracing_v3/__init__.py +0 -10
  1042. synth_ai/v0/tracing_v3/abstractions.py +0 -3
  1043. synth_ai/v0/tracing_v3/decorators.py +0 -3
  1044. synth_ai/v0/tracing_v3/llm_call_record_helpers.py +0 -3
  1045. synth_ai/v0/tracing_v3/session_tracer.py +0 -3
  1046. synth_ai-0.2.14.dist-info/METADATA +0 -139
  1047. synth_ai-0.2.14.dist-info/RECORD +0 -762
  1048. synth_ai-0.2.14.dist-info/top_level.txt +0 -2
  1049. /synth_ai/{demos/demo_task_apps → cli/demo_apps}/crafter/__init__.py +0 -0
  1050. /synth_ai/{demos → cli/demo_apps}/demo_task_apps/__init__.py +0 -0
  1051. /synth_ai/{demos → cli/demo_apps}/demo_task_apps/crafter/configs/crafter_fft_4b.toml +0 -0
  1052. /synth_ai/{demos → cli/demo_apps}/demo_task_apps/crafter/configs/rl_from_base_qwen4b.toml +0 -0
  1053. /synth_ai/{demos → cli/demo_apps}/demo_task_apps/math/__init__.py +0 -0
  1054. /synth_ai/{demos → cli/demo_apps}/demo_task_apps/math/_common.py +0 -0
  1055. /synth_ai/{demos → cli/demo_apps}/demo_task_apps/math/app.py +0 -0
  1056. /synth_ai/{demos → cli/demo_apps}/demo_task_apps/math/deploy_modal.py +0 -0
  1057. {examples/task_apps → synth_ai/core/apps}/__init__.py +0 -0
  1058. /synth_ai/{tracing_v3 → core/tracing_v3}/examples/basic_usage.py +0 -0
  1059. /synth_ai/{tracing_v3 → core/tracing_v3}/hooks.py +0 -0
  1060. /synth_ai/{tracing_v3 → core/tracing_v3}/lm_call_record_abstractions.py +0 -0
  1061. /synth_ai/{tracing_v3 → core/tracing_v3}/replica_sync.py +0 -0
  1062. /synth_ai/{tracing_v3 → core/tracing_v3}/serialization.py +0 -0
  1063. /synth_ai/{tracing_v3 → core/tracing_v3}/storage/__init__.py +0 -0
  1064. /synth_ai/{tracing_v3 → core/tracing_v3}/storage/exceptions.py +0 -0
  1065. /synth_ai/{tracing_v3 → core/tracing_v3}/storage/types.py +0 -0
  1066. /synth_ai/{tracing_v3 → core/tracing_v3}/storage/utils.py +0 -0
  1067. /synth_ai/{tracing_v3 → core/tracing_v3}/turso/__init__.py +0 -0
  1068. /synth_ai/{learning → sdk/learning}/algorithms.py +0 -0
  1069. /synth_ai/{learning → sdk/learning}/config.py +0 -0
  1070. /synth_ai/{learning → sdk/learning}/constants.py +0 -0
  1071. /synth_ai/{learning → sdk/learning}/core.py +0 -0
  1072. /synth_ai/{learning → sdk/learning}/gateway.py +0 -0
  1073. /synth_ai/{learning → sdk/learning}/rl/config.py +0 -0
  1074. /synth_ai/{learning → sdk/learning}/rl_client.py +0 -0
  1075. /synth_ai/{learning → sdk/learning}/sft/__init__.py +0 -0
  1076. /synth_ai/{learning → sdk/learning}/sse.py +0 -0
  1077. /synth_ai/{task → sdk/task}/auth.py +0 -0
  1078. /synth_ai/{task → sdk/task}/errors.py +0 -0
  1079. /synth_ai/{task → sdk/task}/health.py +0 -0
  1080. /synth_ai/{task → sdk/task}/json.py +0 -0
  1081. /synth_ai/{task → sdk/task}/rubrics/models.py +0 -0
  1082. /synth_ai/{task → sdk/task}/rubrics/scoring.py +0 -0
  1083. /synth_ai/{task → sdk/task}/vendors.py +0 -0
  1084. {synth_ai-0.2.14.dist-info → synth_ai-0.4.4.dist-info}/WHEEL +0 -0
  1085. {synth_ai-0.2.14.dist-info → synth_ai-0.4.4.dist-info}/entry_points.txt +0 -0
  1086. {synth_ai-0.2.14.dist-info → synth_ai-0.4.4.dist-info}/licenses/LICENSE +0 -0
@@ -1,1209 +0,0 @@
1
- """LibSQL-native trace manager prototype.
2
-
3
- This module provides the Turso/libsql-backed trace storage implementation. It
4
- mirrors the public surface area of the historical SQLAlchemy manager while
5
- executing all operations directly via libsql.
6
- """
7
-
8
- from __future__ import annotations
9
-
10
- import asyncio
11
- import json
12
- import logging
13
- import re
14
- from collections.abc import Callable
15
- from dataclasses import asdict, dataclass
16
- from datetime import datetime, timezone
17
- from typing import TYPE_CHECKING, Any, cast
18
-
19
- import libsql
20
- from sqlalchemy.engine import make_url
21
-
22
- from ..abstractions import (
23
- EnvironmentEvent,
24
- LMCAISEvent,
25
- RuntimeEvent,
26
- SessionMessageContent,
27
- SessionTrace,
28
- )
29
- from ..config import CONFIG
30
- from ..storage.base import TraceStorage
31
- from .models import analytics_views
32
-
33
- if TYPE_CHECKING:
34
- from sqlite3 import Connection as LibsqlConnection
35
- else: # pragma: no cover - runtime fallback for typing only
36
- LibsqlConnection = Any # type: ignore[assignment]
37
-
38
- _LIBSQL_CONNECT_ATTR = getattr(libsql, "connect", None)
39
- if _LIBSQL_CONNECT_ATTR is None: # pragma: no cover - defensive guard
40
- raise RuntimeError("libsql.connect is required for NativeLibsqlTraceManager")
41
- _libsql_connect: Callable[..., LibsqlConnection] = cast(
42
- Callable[..., LibsqlConnection],
43
- _LIBSQL_CONNECT_ATTR,
44
- )
45
-
46
- try: # pragma: no cover - exercised only when pandas present
47
- import pandas as pd # type: ignore
48
- except Exception: # pragma: no cover
49
- pd = None # type: ignore[assignment]
50
-
51
- logger = logging.getLogger(__name__)
52
-
53
-
54
- @dataclass(slots=True)
55
- class _ConnectionTarget:
56
- """Resolved connection target for libsql."""
57
-
58
- database: str
59
- sync_url: str | None = None
60
- auth_token: str | None = None
61
-
62
-
63
- def _resolve_connection_target(db_url: str | None, auth_token: str | None) -> _ConnectionTarget:
64
- """Normalise the configured database URL."""
65
- url = db_url or CONFIG.db_url
66
-
67
- # Fast-path local SQLite URLs (`sqlite+aiosqlite:///path/to/db`)
68
- if url.startswith("sqlite+aiosqlite:///"):
69
- return _ConnectionTarget(database=url.replace("sqlite+aiosqlite:///", ""), auth_token=auth_token)
70
-
71
- # SQLAlchemy-compatible libsql scheme (`sqlite+libsql://<endpoint or path>`)
72
- if url.startswith("sqlite+libsql://"):
73
- target = url.replace("sqlite+libsql://", "", 1)
74
- return _ConnectionTarget(database=target, sync_url=target if target.startswith("libsql://") else None, auth_token=auth_token)
75
-
76
- # Native libsql URLs (`libsql://...`).
77
- if url.startswith("libsql://"):
78
- return _ConnectionTarget(database=url, sync_url=url, auth_token=auth_token)
79
-
80
- # Fallback to SQLAlchemy URL parsing for anything else we missed.
81
- try:
82
- parsed = make_url(url)
83
- if parsed.drivername.startswith("sqlite") and parsed.database:
84
- return _ConnectionTarget(database=parsed.database, auth_token=auth_token)
85
- if parsed.drivername.startswith("libsql"):
86
- database = parsed.render_as_string(hide_password=False)
87
- return _ConnectionTarget(database=database, sync_url=database, auth_token=auth_token)
88
- except Exception: # pragma: no cover - defensive guardrail
89
- logger.debug("Unable to parse db_url via SQLAlchemy", exc_info=True)
90
-
91
- # As a last resort use the raw value (libsql.connect can handle absolute paths).
92
- return _ConnectionTarget(database=url, auth_token=auth_token)
93
-
94
-
95
- def _json_dumps(value: Any) -> str | None:
96
- """Serialise Python objects as JSON compatible with the existing schema."""
97
-
98
- def _default(obj: Any):
99
- if isinstance(obj, datetime):
100
- return obj.isoformat()
101
- return str(obj)
102
-
103
- if value is None:
104
- return None
105
- return json.dumps(value, separators=(",", ":"), default=_default)
106
-
107
-
108
- def _maybe_datetime(value: Any) -> Any:
109
- if value is None or isinstance(value, datetime):
110
- return value
111
- if isinstance(value, str):
112
- try:
113
- return datetime.fromisoformat(value)
114
- except ValueError:
115
- pass
116
- return value
117
-
118
-
119
- def _load_json(value: Any) -> Any:
120
- if value is None or isinstance(value, dict | list):
121
- return value or {}
122
- if isinstance(value, str):
123
- try:
124
- return json.loads(value)
125
- except (TypeError, ValueError):
126
- return {}
127
- return value
128
-
129
-
130
- _TABLE_DEFINITIONS: tuple[str, ...] = (
131
- """
132
- CREATE TABLE IF NOT EXISTS experiments (
133
- experiment_id VARCHAR PRIMARY KEY,
134
- name VARCHAR NOT NULL,
135
- description TEXT,
136
- created_at DATETIME DEFAULT CURRENT_TIMESTAMP,
137
- updated_at DATETIME DEFAULT CURRENT_TIMESTAMP,
138
- configuration TEXT,
139
- metadata TEXT
140
- )
141
- """,
142
- """
143
- CREATE TABLE IF NOT EXISTS systems (
144
- system_id VARCHAR PRIMARY KEY,
145
- name VARCHAR NOT NULL,
146
- system_type VARCHAR,
147
- description TEXT,
148
- created_at DATETIME DEFAULT CURRENT_TIMESTAMP,
149
- metadata TEXT
150
- )
151
- """,
152
- """
153
- CREATE TABLE IF NOT EXISTS system_versions (
154
- version_id VARCHAR PRIMARY KEY,
155
- system_id VARCHAR NOT NULL,
156
- version_number VARCHAR NOT NULL,
157
- commit_hash VARCHAR,
158
- created_at DATETIME DEFAULT CURRENT_TIMESTAMP,
159
- configuration TEXT,
160
- metadata TEXT,
161
- FOREIGN KEY(system_id) REFERENCES systems(system_id),
162
- UNIQUE(system_id, version_number)
163
- )
164
- """,
165
- """
166
- CREATE TABLE IF NOT EXISTS experimental_systems (
167
- id INTEGER PRIMARY KEY AUTOINCREMENT,
168
- experiment_id VARCHAR NOT NULL,
169
- system_id VARCHAR NOT NULL,
170
- version_id VARCHAR NOT NULL,
171
- FOREIGN KEY(experiment_id) REFERENCES experiments(experiment_id),
172
- FOREIGN KEY(system_id) REFERENCES systems(system_id),
173
- FOREIGN KEY(version_id) REFERENCES system_versions(version_id)
174
- )
175
- """,
176
- """
177
- CREATE TABLE IF NOT EXISTS session_traces (
178
- session_id VARCHAR PRIMARY KEY,
179
- created_at DATETIME NOT NULL,
180
- num_timesteps INTEGER NOT NULL,
181
- num_events INTEGER NOT NULL,
182
- num_messages INTEGER NOT NULL,
183
- metadata TEXT,
184
- experiment_id VARCHAR,
185
- embedding VECTOR,
186
- FOREIGN KEY(experiment_id) REFERENCES experiments(experiment_id)
187
- )
188
- """,
189
- """
190
- CREATE TABLE IF NOT EXISTS session_timesteps (
191
- id INTEGER PRIMARY KEY AUTOINCREMENT,
192
- session_id VARCHAR NOT NULL,
193
- step_id VARCHAR NOT NULL,
194
- step_index INTEGER NOT NULL,
195
- turn_number INTEGER,
196
- started_at DATETIME,
197
- completed_at DATETIME,
198
- num_events INTEGER,
199
- num_messages INTEGER,
200
- step_metadata TEXT,
201
- UNIQUE(session_id, step_id),
202
- FOREIGN KEY(session_id) REFERENCES session_traces(session_id)
203
- )
204
- """,
205
- """
206
- CREATE TABLE IF NOT EXISTS events (
207
- id INTEGER PRIMARY KEY AUTOINCREMENT,
208
- session_id VARCHAR NOT NULL,
209
- timestep_id INTEGER,
210
- event_type VARCHAR NOT NULL,
211
- system_instance_id VARCHAR,
212
- event_time FLOAT,
213
- message_time INTEGER,
214
- created_at DATETIME,
215
- model_name VARCHAR,
216
- provider VARCHAR,
217
- input_tokens INTEGER,
218
- output_tokens INTEGER,
219
- total_tokens INTEGER,
220
- cost_usd INTEGER,
221
- latency_ms INTEGER,
222
- span_id VARCHAR,
223
- trace_id VARCHAR,
224
- call_records TEXT,
225
- reward FLOAT,
226
- terminated BOOLEAN,
227
- truncated BOOLEAN,
228
- system_state_before TEXT,
229
- system_state_after TEXT,
230
- metadata TEXT,
231
- event_metadata TEXT,
232
- embedding VECTOR,
233
- CHECK (event_type IN ('cais', 'environment', 'runtime')),
234
- FOREIGN KEY(session_id) REFERENCES session_traces(session_id),
235
- FOREIGN KEY(timestep_id) REFERENCES session_timesteps(id)
236
- )
237
- """,
238
- """
239
- CREATE TABLE IF NOT EXISTS messages (
240
- id INTEGER PRIMARY KEY AUTOINCREMENT,
241
- session_id VARCHAR NOT NULL,
242
- timestep_id INTEGER,
243
- message_type VARCHAR NOT NULL,
244
- content TEXT NOT NULL,
245
- timestamp DATETIME,
246
- event_time FLOAT,
247
- message_time INTEGER,
248
- metadata TEXT,
249
- embedding VECTOR,
250
- CHECK (message_type IN ('user', 'assistant', 'system', 'tool_use', 'tool_result')),
251
- FOREIGN KEY(session_id) REFERENCES session_traces(session_id),
252
- FOREIGN KEY(timestep_id) REFERENCES session_timesteps(id)
253
- )
254
- """,
255
- """
256
- CREATE TABLE IF NOT EXISTS outcome_rewards (
257
- id INTEGER PRIMARY KEY AUTOINCREMENT,
258
- session_id VARCHAR NOT NULL,
259
- total_reward INTEGER NOT NULL,
260
- achievements_count INTEGER NOT NULL,
261
- total_steps INTEGER NOT NULL,
262
- created_at DATETIME NOT NULL,
263
- reward_metadata TEXT,
264
- FOREIGN KEY(session_id) REFERENCES session_traces(session_id)
265
- )
266
- """,
267
- """
268
- CREATE TABLE IF NOT EXISTS event_rewards (
269
- id INTEGER PRIMARY KEY AUTOINCREMENT,
270
- event_id INTEGER NOT NULL,
271
- session_id VARCHAR NOT NULL,
272
- message_id INTEGER,
273
- turn_number INTEGER,
274
- reward_value FLOAT NOT NULL,
275
- reward_type VARCHAR,
276
- "key" VARCHAR,
277
- annotation TEXT,
278
- source VARCHAR,
279
- created_at DATETIME NOT NULL,
280
- FOREIGN KEY(event_id) REFERENCES events(id),
281
- FOREIGN KEY(session_id) REFERENCES session_traces(session_id),
282
- FOREIGN KEY(message_id) REFERENCES messages(id)
283
- )
284
- """
285
- )
286
-
287
-
288
- _INDEX_DEFINITIONS: tuple[str, ...] = (
289
- "CREATE INDEX IF NOT EXISTS idx_session_created ON session_traces (created_at)",
290
- "CREATE INDEX IF NOT EXISTS idx_session_experiment ON session_traces (experiment_id)",
291
- "CREATE INDEX IF NOT EXISTS idx_timestep_session_step ON session_timesteps (session_id, step_id)",
292
- "CREATE INDEX IF NOT EXISTS idx_timestep_turn ON session_timesteps (turn_number)",
293
- "CREATE INDEX IF NOT EXISTS idx_event_session_step ON events (session_id, timestep_id)",
294
- "CREATE INDEX IF NOT EXISTS idx_event_type ON events (event_type)",
295
- "CREATE INDEX IF NOT EXISTS idx_event_created ON events (created_at)",
296
- "CREATE INDEX IF NOT EXISTS idx_event_model ON events (model_name)",
297
- "CREATE INDEX IF NOT EXISTS idx_event_trace ON events (trace_id)",
298
- "CREATE INDEX IF NOT EXISTS idx_message_session_step ON messages (session_id, timestep_id)",
299
- "CREATE INDEX IF NOT EXISTS idx_message_type ON messages (message_type)",
300
- "CREATE INDEX IF NOT EXISTS idx_message_timestamp ON messages (timestamp)",
301
- "CREATE INDEX IF NOT EXISTS idx_experiment_created ON experiments (created_at)",
302
- "CREATE INDEX IF NOT EXISTS idx_experiment_name ON experiments (name)",
303
- "CREATE INDEX IF NOT EXISTS idx_system_name ON systems (name)",
304
- "CREATE INDEX IF NOT EXISTS idx_system_type ON systems (system_type)",
305
- "CREATE UNIQUE INDEX IF NOT EXISTS uq_system_version ON system_versions (system_id, version_number)",
306
- "CREATE INDEX IF NOT EXISTS idx_version_system ON system_versions (system_id)",
307
- "CREATE INDEX IF NOT EXISTS idx_version_created ON system_versions (created_at)",
308
- "CREATE UNIQUE INDEX IF NOT EXISTS uq_experiment_system ON experimental_systems (experiment_id, system_id)",
309
- "CREATE INDEX IF NOT EXISTS idx_experimental_system ON experimental_systems (experiment_id, system_id)",
310
- "CREATE INDEX IF NOT EXISTS idx_outcome_rewards_session ON outcome_rewards (session_id)",
311
- "CREATE INDEX IF NOT EXISTS idx_outcome_rewards_total ON outcome_rewards (total_reward)",
312
- "CREATE INDEX IF NOT EXISTS idx_event_rewards_session ON event_rewards (session_id)",
313
- "CREATE INDEX IF NOT EXISTS idx_event_rewards_event ON event_rewards (event_id)",
314
- "CREATE INDEX IF NOT EXISTS idx_event_rewards_type ON event_rewards (reward_type)",
315
- 'CREATE INDEX IF NOT EXISTS idx_event_rewards_key ON event_rewards ("key")',
316
- )
317
-
318
-
319
- class NativeLibsqlTraceManager(TraceStorage):
320
- """Libsql-backed trace manager."""
321
-
322
- def __init__(
323
- self,
324
- db_url: str | None = None,
325
- *,
326
- auth_token: str | None = None,
327
- ):
328
- self._config_auth_token = auth_token
329
- self._target = _resolve_connection_target(db_url, auth_token)
330
- self._conn: LibsqlConnection | None = None
331
- self._conn_lock = asyncio.Lock()
332
- self._op_lock = asyncio.Lock()
333
- self._initialized = False
334
-
335
- def _open_connection(self) -> LibsqlConnection:
336
- """Open a libsql connection for the resolved target."""
337
- kwargs: dict[str, Any] = {}
338
- if self._target.sync_url and self._target.sync_url.startswith("libsql://"):
339
- kwargs["sync_url"] = self._target.sync_url
340
- if self._target.auth_token:
341
- kwargs["auth_token"] = self._target.auth_token
342
- # Disable automatic background sync; ReplicaSync drives this explicitly.
343
- kwargs.setdefault("sync_interval", 0)
344
- logger.debug("Opening libsql connection to %s", self._target.database)
345
- return _libsql_connect(self._target.database, **kwargs)
346
-
347
- async def initialize(self):
348
- """Initialise the backend."""
349
- async with self._conn_lock:
350
- if self._initialized:
351
- return
352
-
353
- # Establish a libsql connection for future native operations.
354
- self._conn = self._open_connection()
355
- self._ensure_schema()
356
- self._initialized = True
357
-
358
- async def close(self):
359
- """Close the libsql connection."""
360
- async with self._conn_lock:
361
- if self._conn:
362
- logger.debug("Closing libsql connection to %s", self._target.database)
363
- self._conn.close()
364
- self._conn = None
365
- self._initialized = False
366
-
367
- # ------------------------------------------------------------------
368
- # Delegated operations (to be swapped with native libsql versions).
369
- # ------------------------------------------------------------------
370
-
371
- async def insert_session_trace(self, trace: SessionTrace) -> str:
372
- await self.initialize()
373
-
374
- import logging as _logging
375
- _logger = _logging.getLogger(__name__)
376
- _logger.info(f"[TRACE_DEBUG] insert_session_trace START: session_id={trace.session_id}, {len(trace.markov_blanket_message_history)} messages")
377
-
378
- session_exists = await self._session_exists(trace.session_id)
379
- _logger.info(f"[TRACE_DEBUG] Session exists: {session_exists}")
380
-
381
- if session_exists:
382
- _logger.warning(f"[TRACE_DEBUG] Session {trace.session_id} already exists, need to save messages anyway!")
383
- # Don't return early - we need to save messages!
384
- # Just update metadata
385
- async with self._op_lock:
386
- conn = self._conn
387
- assert conn is not None
388
- conn.execute(
389
- "UPDATE session_traces SET metadata = ? WHERE session_id = ?",
390
- (_json_dumps(trace.metadata or {}), trace.session_id),
391
- )
392
- conn.commit()
393
- # Continue to save messages instead of returning
394
-
395
- if not session_exists:
396
- created_at = trace.created_at or datetime.now(timezone.utc)
397
-
398
- async with self._op_lock:
399
- conn = self._conn
400
- assert conn is not None
401
- conn.execute(
402
- """
403
- INSERT INTO session_traces (
404
- session_id,
405
- created_at,
406
- num_timesteps,
407
- num_events,
408
- num_messages,
409
- metadata
410
- )
411
- VALUES (?, ?, 0, 0, 0, ?)
412
- """,
413
- (
414
- trace.session_id,
415
- created_at.isoformat(),
416
- _json_dumps(trace.metadata or {}),
417
- ),
418
- )
419
- conn.commit()
420
- _logger.info(f"[TRACE_DEBUG] Session row inserted")
421
-
422
- step_id_map: dict[str, int] = {}
423
-
424
- for step in trace.session_time_steps:
425
- step_db_id = await self.ensure_timestep(
426
- trace.session_id,
427
- step_id=step.step_id,
428
- step_index=step.step_index,
429
- turn_number=step.turn_number,
430
- started_at=step.timestamp,
431
- completed_at=step.completed_at,
432
- metadata=step.step_metadata or {},
433
- )
434
- step_id_map[step.step_id] = step_db_id
435
-
436
- for event in trace.event_history:
437
- step_ref = None
438
- metadata = event.metadata or {}
439
- if isinstance(metadata, dict):
440
- step_ref = metadata.get("step_id")
441
- timestep_db_id = step_id_map.get(step_ref) if step_ref else None
442
- await self.insert_event_row(
443
- trace.session_id,
444
- timestep_db_id=timestep_db_id,
445
- event=event,
446
- metadata_override=event.metadata or {},
447
- )
448
-
449
- import logging as _logging
450
- _logger = _logging.getLogger(__name__)
451
- _logger.info(f"[TRACE_DEBUG] insert_session_trace: saving {len(trace.markov_blanket_message_history)} messages")
452
-
453
- for idx, msg in enumerate(trace.markov_blanket_message_history):
454
- metadata = dict(getattr(msg, "metadata", {}) or {})
455
- step_ref = metadata.get("step_id")
456
- content_value = msg.content
457
- if isinstance(msg.content, SessionMessageContent):
458
- if msg.content.json_payload:
459
- metadata.setdefault("json_payload", msg.content.json_payload)
460
- content_value = msg.content.json_payload
461
- else:
462
- content_value = msg.content.as_text()
463
- if msg.content.text:
464
- metadata.setdefault("text", msg.content.text)
465
- elif not isinstance(content_value, str):
466
- try:
467
- content_value = json.dumps(content_value, ensure_ascii=False)
468
- except (TypeError, ValueError):
469
- content_value = str(content_value)
470
-
471
- _logger.info(f"[TRACE_DEBUG] Message {idx+1}: type={msg.message_type}, content_len={len(str(content_value))}")
472
-
473
- try:
474
- await self.insert_message_row(
475
- trace.session_id,
476
- timestep_db_id=step_id_map.get(step_ref) if step_ref else None,
477
- message_type=msg.message_type,
478
- content=content_value,
479
- event_time=msg.time_record.event_time,
480
- message_time=msg.time_record.message_time,
481
- metadata=metadata,
482
- )
483
- _logger.info(f"[TRACE_DEBUG] Message {idx+1}: saved successfully")
484
- except Exception as exc:
485
- _logger.error(f"[TRACE_DEBUG] Message {idx+1}: FAILED TO SAVE: {exc}", exc_info=True)
486
- raise
487
-
488
- async with self._op_lock:
489
- conn = self._conn
490
- assert conn is not None
491
- conn.execute(
492
- "UPDATE session_traces SET num_timesteps = ?, num_events = ?, num_messages = ?, metadata = ? WHERE session_id = ?",
493
- (
494
- len(trace.session_time_steps),
495
- len(trace.event_history),
496
- len(trace.markov_blanket_message_history),
497
- _json_dumps(trace.metadata or {}),
498
- trace.session_id,
499
- ),
500
- )
501
- conn.commit()
502
-
503
- return trace.session_id
504
-
505
- async def get_session_trace(self, session_id: str) -> dict[str, Any] | None:
506
- await self.initialize()
507
-
508
- async with self._op_lock:
509
- conn = self._conn
510
- assert conn is not None
511
-
512
- session_cursor = conn.execute(
513
- """
514
- SELECT session_id,
515
- created_at,
516
- num_timesteps,
517
- num_events,
518
- num_messages,
519
- metadata
520
- FROM session_traces
521
- WHERE session_id = ?
522
- """,
523
- (session_id,),
524
- )
525
- session_row = session_cursor.fetchone()
526
- session_cursor.close()
527
-
528
- if not session_row:
529
- return None
530
-
531
- session_columns = ["session_id", "created_at", "num_timesteps", "num_events", "num_messages", "metadata"]
532
- session_data = dict(zip(session_columns, session_row, strict=True))
533
-
534
- timestep_cursor = conn.execute(
535
- """
536
- SELECT step_id,
537
- step_index,
538
- turn_number,
539
- started_at,
540
- completed_at,
541
- step_metadata
542
- FROM session_timesteps
543
- WHERE session_id = ?
544
- ORDER BY step_index ASC
545
- """,
546
- (session_id,),
547
- )
548
- timestep_rows = timestep_cursor.fetchall()
549
- timestep_cursor.close()
550
-
551
- return {
552
- "session_id": session_data["session_id"],
553
- "created_at": _maybe_datetime(session_data["created_at"]),
554
- "num_timesteps": session_data["num_timesteps"],
555
- "num_events": session_data["num_events"],
556
- "num_messages": session_data["num_messages"],
557
- "metadata": _load_json(session_data["metadata"]),
558
- "timesteps": [
559
- {
560
- "step_id": row[0],
561
- "step_index": row[1],
562
- "turn_number": row[2],
563
- "started_at": _maybe_datetime(row[3]),
564
- "completed_at": _maybe_datetime(row[4]),
565
- "metadata": _load_json(row[5]),
566
- }
567
- for row in timestep_rows
568
- ],
569
- }
570
-
571
- async def _session_exists(self, session_id: str) -> bool:
572
- await self.initialize()
573
- async with self._op_lock:
574
- conn = self._conn
575
- assert conn is not None
576
- cursor = conn.execute(
577
- "SELECT 1 FROM session_traces WHERE session_id = ?", (session_id,)
578
- )
579
- row = cursor.fetchone()
580
- cursor.close()
581
- return row is not None
582
-
583
- @staticmethod
584
- def _normalise_params(params: dict[str, Any] | None) -> dict[str, Any]:
585
- if not params:
586
- return {}
587
- normalised: dict[str, Any] = {}
588
- for key, value in params.items():
589
- if isinstance(value, datetime):
590
- normalised[key] = value.isoformat()
591
- else:
592
- normalised[key] = value
593
- return normalised
594
-
595
- @staticmethod
596
- def _prepare_query_params(query: str, params: dict[str, Any] | list[Any] | tuple[Any, ...]) -> tuple[str, tuple[Any, ...]]:
597
- if isinstance(params, dict):
598
- keys: list[str] = []
599
-
600
- def _replace(match: re.Match[str]) -> str:
601
- key = match.group(1)
602
- keys.append(key)
603
- return "?"
604
-
605
- new_query = re.sub(r":([a-zA-Z_][a-zA-Z0-9_]*)", _replace, query)
606
- if not keys:
607
- raise ValueError("No named parameters found in query for provided mapping")
608
- values = tuple(params[key] for key in keys)
609
- return new_query, values
610
- if isinstance(params, list | tuple):
611
- return query, tuple(params)
612
- raise TypeError("Unsupported parameter type for query execution")
613
-
614
- def _ensure_schema(self) -> None:
615
- if not self._conn:
616
- raise RuntimeError("Connection not initialised")
617
-
618
- for ddl in _TABLE_DEFINITIONS:
619
- self._conn.execute(ddl)
620
- for ddl in _INDEX_DEFINITIONS:
621
- self._conn.execute(ddl)
622
- for view_sql in analytics_views.values():
623
- self._conn.execute(view_sql)
624
- self._conn.commit()
625
-
626
- async def query_traces(self, query: str, params: dict[str, Any] | None = None) -> Any:
627
- await self.initialize()
628
-
629
- async with self._op_lock:
630
- conn = self._conn
631
- assert conn is not None
632
- normalised = self._normalise_params(params)
633
- if normalised:
634
- prepared_query, prepared_params = self._prepare_query_params(query, normalised)
635
- cursor = conn.execute(prepared_query, prepared_params)
636
- else:
637
- cursor = conn.execute(query)
638
- try:
639
- description = cursor.description or []
640
- columns = [col[0] for col in description]
641
- rows = cursor.fetchall()
642
- finally:
643
- cursor.close()
644
-
645
- if not rows:
646
- if pd is not None:
647
- return pd.DataFrame(columns=list(columns))
648
- return []
649
-
650
- records = [dict(zip(columns, row, strict=True)) for row in rows]
651
- if pd is not None:
652
- return pd.DataFrame(records)
653
- return records
654
-
655
- async def get_model_usage(
656
- self,
657
- start_date=None,
658
- end_date=None,
659
- model_name=None,
660
- ) -> Any:
661
- query = """
662
- SELECT * FROM model_usage_stats
663
- WHERE 1=1
664
- """
665
- params: dict[str, Any] = {}
666
- if start_date:
667
- params["start_date"] = start_date
668
- query += " AND last_used >= :start_date"
669
- if end_date:
670
- params["end_date"] = end_date
671
- query += " AND first_used <= :end_date"
672
- if model_name:
673
- params["model_name"] = model_name
674
- query += " AND model_name = :model_name"
675
- query += " ORDER BY usage_count DESC"
676
- return await self.query_traces(query, params)
677
-
678
- async def delete_session(self, session_id: str) -> bool:
679
- await self.initialize()
680
-
681
- async with self._op_lock:
682
- conn = self._conn
683
- assert conn is not None
684
-
685
- cursor = conn.execute(
686
- "SELECT 1 FROM session_traces WHERE session_id = ?", (session_id,)
687
- )
688
- exists = cursor.fetchone() is not None
689
- cursor.close()
690
- if not exists:
691
- return False
692
-
693
- conn.execute("DELETE FROM event_rewards WHERE session_id = ?", (session_id,))
694
- conn.execute("DELETE FROM outcome_rewards WHERE session_id = ?", (session_id,))
695
- conn.execute("DELETE FROM messages WHERE session_id = ?", (session_id,))
696
- conn.execute("DELETE FROM events WHERE session_id = ?", (session_id,))
697
- conn.execute("DELETE FROM session_timesteps WHERE session_id = ?", (session_id,))
698
- conn.execute("DELETE FROM session_traces WHERE session_id = ?", (session_id,))
699
- conn.commit()
700
- return True
701
-
702
- # Experiment helpers -------------------------------------------------
703
- async def create_experiment(
704
- self,
705
- experiment_id: str,
706
- name: str,
707
- description: str | None = None,
708
- configuration: dict[str, Any] | None = None,
709
- ) -> str:
710
- await self.initialize()
711
-
712
- async with self._op_lock:
713
- conn = self._conn
714
- assert conn is not None
715
- conn.execute(
716
- """
717
- INSERT INTO experiments (experiment_id, name, description, configuration)
718
- VALUES (?, ?, ?, ?)
719
- ON CONFLICT(experiment_id) DO UPDATE SET
720
- name = excluded.name,
721
- description = excluded.description,
722
- configuration = excluded.configuration
723
- """,
724
- (
725
- experiment_id,
726
- name,
727
- description,
728
- _json_dumps(configuration or {}),
729
- ),
730
- )
731
- conn.commit()
732
- return experiment_id
733
-
734
- async def link_session_to_experiment(self, session_id: str, experiment_id: str):
735
- await self.initialize()
736
-
737
- async with self._op_lock:
738
- conn = self._conn
739
- assert conn is not None
740
- conn.execute(
741
- "UPDATE session_traces SET experiment_id = ? WHERE session_id = ?",
742
- (experiment_id, session_id),
743
- )
744
- conn.commit()
745
-
746
- async def get_sessions_by_experiment(
747
- self, experiment_id: str, limit: int | None = None
748
- ) -> list[dict[str, Any]]:
749
- await self.initialize()
750
-
751
- sql = """
752
- SELECT session_id,
753
- created_at,
754
- num_timesteps,
755
- num_events,
756
- num_messages,
757
- metadata
758
- FROM session_traces
759
- WHERE experiment_id = ?
760
- ORDER BY created_at DESC
761
- """
762
- params: list[Any] = [experiment_id]
763
- if limit is not None:
764
- sql += " LIMIT ?"
765
- params.append(limit)
766
-
767
- async with self._op_lock:
768
- conn = self._conn
769
- assert conn is not None
770
- cursor = conn.execute(sql, params)
771
- rows = cursor.fetchall()
772
- cursor.close()
773
-
774
- return [
775
- {
776
- "session_id": row[0],
777
- "created_at": _maybe_datetime(row[1]),
778
- "num_timesteps": row[2],
779
- "num_events": row[3],
780
- "num_messages": row[4],
781
- "metadata": _load_json(row[5]),
782
- }
783
- for row in rows
784
- ]
785
-
786
- async def batch_insert_sessions(
787
- self, traces: list[SessionTrace], batch_size: int | None = None
788
- ) -> list[str]:
789
- batch_size = batch_size or CONFIG.batch_size
790
- inserted: list[str] = []
791
-
792
- for i in range(0, len(traces), batch_size):
793
- chunk = traces[i : i + batch_size]
794
- for trace in chunk:
795
- session_id = await self.insert_session_trace(trace)
796
- inserted.append(session_id)
797
- return inserted
798
-
799
- # Incremental helpers -----------------------------------------------
800
- async def ensure_session(
801
- self,
802
- session_id: str,
803
- *,
804
- created_at=None,
805
- metadata=None,
806
- ) -> None:
807
- await self.initialize()
808
-
809
- created_at_val = (created_at or datetime.now(timezone.utc)).isoformat()
810
- metadata_json = _json_dumps(metadata or {})
811
-
812
- async with self._op_lock:
813
- conn = self._conn
814
-
815
- assert conn is not None
816
- conn.execute(
817
- """
818
- INSERT INTO session_traces (
819
- session_id, created_at, num_timesteps, num_events, num_messages, metadata
820
- )
821
- VALUES (?, ?, 0, 0, 0, ?)
822
- ON CONFLICT(session_id) DO NOTHING
823
- """,
824
- (session_id, created_at_val, metadata_json),
825
- )
826
- conn.commit()
827
-
828
- async def ensure_timestep(
829
- self,
830
- session_id: str,
831
- *,
832
- step_id: str,
833
- step_index: int,
834
- turn_number: int | None = None,
835
- started_at=None,
836
- completed_at=None,
837
- metadata=None,
838
- ) -> int:
839
- await self.initialize()
840
-
841
- started_at_val = (started_at or datetime.now(timezone.utc)).isoformat()
842
- completed_at_val = completed_at.isoformat() if completed_at else None
843
- metadata_json = _json_dumps(metadata or {})
844
-
845
- async with self._op_lock:
846
- conn = self._conn
847
-
848
- assert conn is not None
849
- cur = conn.execute(
850
- """
851
- SELECT id FROM session_timesteps
852
- WHERE session_id = ? AND step_id = ?
853
- """,
854
- (session_id, step_id),
855
- )
856
- row = cur.fetchone()
857
- if row:
858
- return int(row[0])
859
-
860
- cur = conn.execute(
861
- """
862
- INSERT INTO session_timesteps (
863
- session_id,
864
- step_id,
865
- step_index,
866
- turn_number,
867
- started_at,
868
- completed_at,
869
- num_events,
870
- num_messages,
871
- step_metadata
872
- )
873
- VALUES (?, ?, ?, ?, ?, ?, 0, 0, ?)
874
- """,
875
- (
876
- session_id,
877
- step_id,
878
- step_index,
879
- turn_number,
880
- started_at_val,
881
- completed_at_val,
882
- metadata_json,
883
- ),
884
- )
885
- timestep_id = int(cur.lastrowid)
886
- conn.execute(
887
- """
888
- UPDATE session_traces
889
- SET num_timesteps = num_timesteps + 1
890
- WHERE session_id = ?
891
- """,
892
- (session_id,),
893
- )
894
- conn.commit()
895
- return timestep_id
896
-
897
- async def insert_event_row(
898
- self,
899
- session_id: str,
900
- *,
901
- timestep_db_id: int | None,
902
- event: Any,
903
- metadata_override: dict[str, Any] | None = None,
904
- ) -> int:
905
- await self.initialize()
906
-
907
- if not isinstance(event, EnvironmentEvent | LMCAISEvent | RuntimeEvent):
908
- raise TypeError(f"Unsupported event type for native manager: {type(event)!r}")
909
-
910
- metadata_json = metadata_override or event.metadata or {}
911
- event_extra_metadata = getattr(event, "event_metadata", None)
912
- system_state_before = getattr(event, "system_state_before", None)
913
- system_state_after = getattr(event, "system_state_after", None)
914
-
915
- payload: dict[str, Any] = {
916
- "session_id": session_id,
917
- "timestep_id": timestep_db_id,
918
- "system_instance_id": event.system_instance_id,
919
- "event_time": event.time_record.event_time,
920
- "message_time": event.time_record.message_time,
921
- "metadata": metadata_json,
922
- "event_metadata": event_extra_metadata,
923
- "system_state_before": system_state_before,
924
- "system_state_after": system_state_after,
925
- }
926
-
927
- if isinstance(event, LMCAISEvent):
928
- call_records = None
929
- if getattr(event, "call_records", None):
930
- call_records = [asdict(record) for record in event.call_records]
931
- payload.update(
932
- {
933
- "event_type": "cais",
934
- "model_name": event.model_name,
935
- "provider": event.provider,
936
- "input_tokens": event.input_tokens,
937
- "output_tokens": event.output_tokens,
938
- "total_tokens": event.total_tokens,
939
- "cost_usd": int(event.cost_usd * 100) if event.cost_usd is not None else None,
940
- "latency_ms": event.latency_ms,
941
- "span_id": event.span_id,
942
- "trace_id": event.trace_id,
943
- "call_records": call_records,
944
- }
945
- )
946
- elif isinstance(event, EnvironmentEvent):
947
- payload.update(
948
- {
949
- "event_type": "environment",
950
- "reward": event.reward,
951
- "terminated": event.terminated,
952
- "truncated": event.truncated,
953
- }
954
- )
955
- elif isinstance(event, RuntimeEvent):
956
- payload.update(
957
- {
958
- "event_type": "runtime",
959
- "metadata": {**(event.metadata or {}), "actions": event.actions},
960
- }
961
- )
962
-
963
- async with self._op_lock:
964
- conn = self._conn
965
-
966
- assert conn is not None
967
- cur = conn.execute(
968
- """
969
- INSERT INTO events (
970
- session_id,
971
- timestep_id,
972
- event_type,
973
- system_instance_id,
974
- event_time,
975
- message_time,
976
- model_name,
977
- provider,
978
- input_tokens,
979
- output_tokens,
980
- total_tokens,
981
- cost_usd,
982
- latency_ms,
983
- span_id,
984
- trace_id,
985
- call_records,
986
- reward,
987
- terminated,
988
- truncated,
989
- system_state_before,
990
- system_state_after,
991
- metadata,
992
- event_metadata
993
- )
994
- VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
995
- """,
996
- (
997
- payload["session_id"],
998
- payload["timestep_id"],
999
- payload.get("event_type"),
1000
- payload["system_instance_id"],
1001
- payload["event_time"],
1002
- payload["message_time"],
1003
- payload.get("model_name"),
1004
- payload.get("provider"),
1005
- payload.get("input_tokens"),
1006
- payload.get("output_tokens"),
1007
- payload.get("total_tokens"),
1008
- payload.get("cost_usd"),
1009
- payload.get("latency_ms"),
1010
- payload.get("span_id"),
1011
- payload.get("trace_id"),
1012
- _json_dumps(payload.get("call_records")),
1013
- payload.get("reward"),
1014
- payload.get("terminated"),
1015
- payload.get("truncated"),
1016
- _json_dumps(payload.get("system_state_before")),
1017
- _json_dumps(payload.get("system_state_after")),
1018
- _json_dumps(payload.get("metadata")),
1019
- _json_dumps(payload.get("event_metadata")),
1020
- ),
1021
- )
1022
- event_id = int(cur.lastrowid)
1023
- conn.execute(
1024
- """
1025
- UPDATE session_traces
1026
- SET num_events = num_events + 1
1027
- WHERE session_id = ?
1028
- """,
1029
- (session_id,),
1030
- )
1031
- if timestep_db_id is not None:
1032
- conn.execute(
1033
- """
1034
- UPDATE session_timesteps
1035
- SET num_events = num_events + 1
1036
- WHERE id = ?
1037
- """,
1038
- (timestep_db_id,),
1039
- )
1040
- conn.commit()
1041
- return event_id
1042
-
1043
- async def insert_message_row(
1044
- self,
1045
- session_id: str,
1046
- *,
1047
- timestep_db_id: int | None,
1048
- message_type: str,
1049
- content: Any,
1050
- event_time: float | None = None,
1051
- message_time: int | None = None,
1052
- metadata: dict[str, Any] | None = None,
1053
- ) -> int:
1054
- await self.initialize()
1055
-
1056
- metadata_payload = dict(metadata or {})
1057
- if isinstance(content, SessionMessageContent):
1058
- if content.json_payload:
1059
- metadata_payload.setdefault("json_payload", content.json_payload)
1060
- content_value = content.json_payload
1061
- else:
1062
- content_value = content.as_text()
1063
- if content.text:
1064
- metadata_payload.setdefault("text", content.text)
1065
- else:
1066
- content_value = content
1067
- if not isinstance(content_value, str):
1068
- try:
1069
- content_value = json.dumps(content_value, ensure_ascii=False)
1070
- except (TypeError, ValueError):
1071
- content_value = str(content_value)
1072
-
1073
- async with self._op_lock:
1074
- conn = self._conn
1075
-
1076
- assert conn is not None
1077
- cur = conn.execute(
1078
- """
1079
- INSERT INTO messages (
1080
- session_id,
1081
- timestep_id,
1082
- message_type,
1083
- content,
1084
- event_time,
1085
- message_time,
1086
- metadata
1087
- )
1088
- VALUES (?, ?, ?, ?, ?, ?, ?)
1089
- """,
1090
- (
1091
- session_id,
1092
- timestep_db_id,
1093
- message_type,
1094
- content_value,
1095
- event_time,
1096
- message_time,
1097
- _json_dumps(metadata_payload),
1098
- ),
1099
- )
1100
- message_id = int(cur.lastrowid)
1101
- conn.execute(
1102
- """
1103
- UPDATE session_traces
1104
- SET num_messages = num_messages + 1
1105
- WHERE session_id = ?
1106
- """,
1107
- (session_id,),
1108
- )
1109
- if timestep_db_id is not None:
1110
- conn.execute(
1111
- """
1112
- UPDATE session_timesteps
1113
- SET num_messages = num_messages + 1
1114
- WHERE id = ?
1115
- """,
1116
- (timestep_db_id,),
1117
- )
1118
- conn.commit()
1119
- return message_id
1120
-
1121
- async def insert_outcome_reward(
1122
- self,
1123
- session_id: str,
1124
- *,
1125
- total_reward: int,
1126
- achievements_count: int,
1127
- total_steps: int,
1128
- reward_metadata: dict | None = None,
1129
- ) -> int:
1130
- await self.initialize()
1131
-
1132
- async with self._op_lock:
1133
- conn = self._conn
1134
-
1135
- assert conn is not None
1136
- cur = conn.execute(
1137
- """
1138
- INSERT INTO outcome_rewards (
1139
- session_id,
1140
- total_reward,
1141
- achievements_count,
1142
- total_steps,
1143
- created_at,
1144
- reward_metadata
1145
- )
1146
- VALUES (?, ?, ?, ?, ?, ?)
1147
- """,
1148
- (
1149
- session_id,
1150
- total_reward,
1151
- achievements_count,
1152
- total_steps,
1153
- datetime.now(timezone.utc).isoformat(),
1154
- _json_dumps(reward_metadata),
1155
- ),
1156
- )
1157
- conn.commit()
1158
- return int(cur.lastrowid)
1159
-
1160
- async def insert_event_reward(
1161
- self,
1162
- session_id: str,
1163
- *,
1164
- event_id: int,
1165
- message_id: int | None = None,
1166
- turn_number: int | None = None,
1167
- reward_value: float = 0.0,
1168
- reward_type: str | None = None,
1169
- key: str | None = None,
1170
- annotation: dict[str, Any] | None = None,
1171
- source: str | None = None,
1172
- ) -> int:
1173
- await self.initialize()
1174
-
1175
- async with self._op_lock:
1176
- conn = self._conn
1177
-
1178
- assert conn is not None
1179
- cur = conn.execute(
1180
- """
1181
- INSERT INTO event_rewards (
1182
- event_id,
1183
- session_id,
1184
- message_id,
1185
- turn_number,
1186
- reward_value,
1187
- reward_type,
1188
- key,
1189
- annotation,
1190
- source,
1191
- created_at
1192
- )
1193
- VALUES (?, ?, ?, ?, ?, ?, ?, ?, ?, ?)
1194
- """,
1195
- (
1196
- event_id,
1197
- session_id,
1198
- message_id,
1199
- turn_number,
1200
- reward_value,
1201
- reward_type,
1202
- key,
1203
- _json_dumps(annotation),
1204
- source,
1205
- datetime.now(timezone.utc).isoformat(),
1206
- ),
1207
- )
1208
- conn.commit()
1209
- return int(cur.lastrowid)