empathy-framework 5.0.1__py3-none-any.whl → 5.0.3__py3-none-any.whl

{empathy_framework-5.0.1.dist-info → empathy_framework-5.0.3.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: empathy-framework
-Version: 5.0.1
+Version: 5.0.3
 Summary: AI collaboration framework with real LLM agent execution, AskUserQuestion tool integration, Socratic agent generation, progressive tier escalation (70-85% cost savings), meta-orchestration, dynamic agent composition (6 patterns), intelligent caching (85% hit rate), semantic workflow discovery, visual workflow editor, MCP integration for Claude Code, and multi-agent orchestration.
 Author-email: Patrick Roebuck <admin@smartaimemory.com>
 Maintainer-email: Smart-AI-Memory <admin@smartaimemory.com>
@@ -170,7 +170,7 @@ License-File: LICENSE
 Requires-Dist: pydantic<3.0.0,>=2.0.0
 Requires-Dist: typing-extensions<5.0.0,>=4.0.0
 Requires-Dist: python-dotenv<2.0.0,>=1.0.0
-Requires-Dist: structlog<25.0.0,>=23.0.0
+Requires-Dist: structlog<26.0.0,>=23.0.0
 Requires-Dist: defusedxml<1.0.0,>=0.7.0
 Requires-Dist: rich<14.0.0,>=13.0.0
 Requires-Dist: typer<1.0.0,>=0.9.0
@@ -256,7 +256,7 @@ Requires-Dist: pytest-cov<8.0,>=4.0; extra == "dev"
 Requires-Dist: pytest-xdist<4.0,>=3.5.0; extra == "dev"
 Requires-Dist: pytest-testmon<3.0,>=2.1.0; extra == "dev"
 Requires-Dist: pytest-picked<1.0,>=0.5.0; extra == "dev"
-Requires-Dist: black<26.0,>=24.3.0; extra == "dev"
+Requires-Dist: black<27.0,>=24.3.0; extra == "dev"
 Requires-Dist: mypy<2.0,>=1.0; extra == "dev"
 Requires-Dist: ruff<1.0,>=0.1; extra == "dev"
 Requires-Dist: coverage<8.0,>=7.0; extra == "dev"
@@ -341,15 +341,15 @@ Requires-Dist: mkdocs-material<10.0.0,>=9.4.0; extra == "all"
 Requires-Dist: mkdocstrings[python]<1.0.0,>=0.24.0; extra == "all"
 Requires-Dist: mkdocs-with-pdf<1.0.0,>=0.9.3; extra == "all"
 Requires-Dist: pymdown-extensions<11.0,>=10.0; extra == "all"
-Requires-Dist: pytest<9.0,>=7.0; extra == "all"
-Requires-Dist: pytest-asyncio<1.0,>=0.21; extra == "all"
-Requires-Dist: pytest-cov<5.0,>=4.0; extra == "all"
-Requires-Dist: black<26.0,>=24.3.0; extra == "all"
+Requires-Dist: pytest<10.0,>=7.0; extra == "all"
+Requires-Dist: pytest-asyncio<2.0,>=0.21; extra == "all"
+Requires-Dist: pytest-cov<8.0,>=4.0; extra == "all"
+Requires-Dist: black<27.0,>=24.3.0; extra == "all"
 Requires-Dist: mypy<2.0,>=1.0; extra == "all"
 Requires-Dist: ruff<1.0,>=0.1; extra == "all"
 Requires-Dist: coverage<8.0,>=7.0; extra == "all"
 Requires-Dist: bandit<2.0,>=1.7; extra == "all"
-Requires-Dist: pre-commit<4.0,>=3.0; extra == "all"
+Requires-Dist: pre-commit<5.0,>=3.0; extra == "all"
 Requires-Dist: httpx<1.0.0,>=0.27.0; extra == "all"
 Requires-Dist: urllib3<3.0.0,>=2.3.0; extra == "all"
 Requires-Dist: aiohttp<4.0.0,>=3.10.0; extra == "all"
@@ -386,12 +386,56 @@ pip install empathy-framework[developer]
 **Timeline:**
 - ✅ **v4.8.0 (Jan 2026):** Deprecation warnings for OpenAI/Google/Ollama providers
 - ✅ **v5.0.0 (Jan 26, 2026):** Non-Anthropic providers removed (BREAKING - COMPLETE)
-- 🎯 **v5.1.0 (Feb 2026):** Prompt caching enabled by default
+- ✅ **v5.0.2 (Jan 28, 2026):** Cost optimization suite with batch processing and caching monitoring
 
 **Migration Guide:** [docs/CLAUDE_NATIVE.md](docs/CLAUDE_NATIVE.md)
 
 ---
 
+## What's New in v5.0.2
+
+**💰 50% Cost Savings with Batch API** - Process non-urgent tasks asynchronously:
+
+```bash
+empathy batch submit batch_requests.json  # Submit batch job
+empathy batch status msgbatch_abc123      # Check progress
+empathy batch results msgbatch_abc123 output.json  # Download results
+```
+
+Perfect for: log analysis, report generation, bulk classification, test generation
+
+**📊 Precise Token Counting** - >98% accurate cost tracking:
+
+- Integrated Anthropic's `count_tokens()` API for billing-accurate measurements
+- 3-tier fallback: API → tiktoken (local) → heuristic
+- Cache-aware cost calculation (25% write markup, 90% read discount)
+
+**📈 Cache Performance Monitoring** - Track your 20-30% caching savings:
+
+```bash
+empathy cache stats           # Show hit rates and cost savings
+empathy cache stats --verbose # Detailed token metrics
+empathy cache stats --format json  # Machine-readable output
+```
+
+**🧭 Adaptive Routing Analytics** - Intelligent tier recommendations:
+
+```bash
+empathy routing stats <workflow>    # Performance metrics
+empathy routing check --all         # Tier upgrade recommendations
+empathy routing models --provider anthropic  # Compare models
+```
+
+**🔧 Dashboard Fixes** - All 6 agent coordination patterns now operational:
+- Agent heartbeats displaying correctly
+- Event streaming functional
+- Coordination signals working
+- Approval gates operational
+
+[See Full Changelog](CHANGELOG.md#502---2026-01-28) | [Batch API Guide](docs/BATCH_API_GUIDE.md) | [User API Docs](docs/USER_API_DOCUMENTATION.md)
+
+---
+
 ## What's New in v4.9.0
 
 **⚡ 18x Faster Performance** - Massive performance gains through Phase 2 optimizations:

{empathy_framework-5.0.1.dist-info → empathy_framework-5.0.3.dist-info}/RECORD

@@ -1,4 +1,4 @@
-empathy_framework-5.0.1.dist-info/licenses/LICENSE,sha256=IJ9eeI5KSrD5P7alsn7sI_6_1bDihxBA5S4Sen4jf2k,4937
+empathy_framework-5.0.3.dist-info/licenses/LICENSE,sha256=IJ9eeI5KSrD5P7alsn7sI_6_1bDihxBA5S4Sen4jf2k,4937
 empathy_healthcare_plugin/__init__.py,sha256=4NioL1_86UXzkd-QNkQZUSZ8rKTQGSP0TC9VXP32kQs,295
 empathy_healthcare_plugin/monitors/__init__.py,sha256=Udp8qfZR504QAq5_eQjvtIaE7v06Yguc7nuF40KllQc,196
 empathy_healthcare_plugin/monitors/clinical_protocol_monitor.py,sha256=MWE5t8tW9HWZn_SNo-inx8-0nhdTNGhbcB8ZeDWyXa0,11648
@@ -22,7 +22,7 @@ empathy_llm_toolkit/levels.py,sha256=Hu9G83iY-BRyjMJKNoDNYZ9xFD0Y0LIn_XEBc5aRWx0
 empathy_llm_toolkit/pattern_confidence.py,sha256=6plkrdDTb3GcjeNxohJlpeHnNWFRRcj5U76D2hvLSGA,14153
 empathy_llm_toolkit/pattern_resolver.py,sha256=VLTeGvEDfQyDwNy5u-t18A5c0BcCpp1FqdExxfqZSlw,9292
 empathy_llm_toolkit/pattern_summary.py,sha256=gA_rCXgfOT1Kp6FP5f1UZqXPp8_mkYXlAfTxc2lhOnc,12286
-empathy_llm_toolkit/providers.py,sha256=AXu9cdYKeG-aooCCnG2Cm7uc26gk3gV9BFH0aVZ3-aA,28621
+empathy_llm_toolkit/providers.py,sha256=v-HcN-v7CykKZmgzLdn2iyjTK5-dtCTBsdpWauOG24A,34852
 empathy_llm_toolkit/session_status.py,sha256=a8QPYf8Sp7jnzHkeRNOyXqoj0QwwoVwkvxPXW9qdvgc,25722
 empathy_llm_toolkit/state.py,sha256=diLXx8QKzfucVDhbKNWej40JXQFmIeM1Qvvzg92pmjU,8030
 empathy_llm_toolkit/agent_factory/__init__.py,sha256=JlusqsYjl_1FqW75h-Uikw2al6v56FAfGSHmQ6gMOk0,1759
@@ -93,8 +93,8 @@ empathy_llm_toolkit/security/secrets_detector_example.py,sha256=LDMJuINpRI-QtWnC
 empathy_llm_toolkit/security/secure_memdocs.py,sha256=x6won-u3tiBkv5ZPVRpSJKTs1mQTZ79jWAEkT8iMxtc,40230
 empathy_llm_toolkit/security/secure_memdocs_example.py,sha256=4UwbaM5X-c93p4_3pz6LKaeNYzfxptzbsYY_-ZuG0A0,8359
 empathy_llm_toolkit/utils/__init__.py,sha256=b2Juy_nuMJ7Ax9tck725V7i3s5shaMSfqZZbiJ2-rOc,187
-empathy_llm_toolkit/utils/tokens.py,sha256=0HwENtc1ud543NzFNeDg3ghH8tGQmNMVUw-LZ93U1YQ,7187
-empathy_os/__init__.py,sha256=R6yG1mUkoaQbw28jF5et5WlNMZxtXEL0nY5yU7SDQaU,12061
+empathy_llm_toolkit/utils/tokens.py,sha256=vu5aggQYmoESIG1dkFahfX6hs6QiyPPF0gMTEp9mGak,11220
+empathy_os/__init__.py,sha256=1k9dqEd8YT2IzC6SYPiq1tCWtdLz3ahC_zGL6fdV3Bw,12061
 empathy_os/agent_monitoring.py,sha256=s4seLC_J4AtQ3PYWrRPO8YHM-Fbm0Q36kPEdlTHf2HI,13375
 empathy_os/cache_monitor.py,sha256=lcBqODhYY9iPaH35PWkOSgyMavdvXneHv9F57dPmjnc,11190
 empathy_os/cache_stats.py,sha256=rWJPBNFEfhuLwKYKy89D_Qa9GPIyVso2jdCKp_cdJhI,10232
@@ -122,7 +122,6 @@ empathy_os/redis_memory.py,sha256=MYhW_M41AoWhxxh8YggWUHxZaUoQZ7XxqeBB2qDqyww,23
 empathy_os/templates.py,sha256=B5fzlc6wU3lwdaKK4ZUdytqOYrM6TEnFteZ0nCSp_EI,17220
 empathy_os/tier_recommender.py,sha256=oYOKhnrTG2sqYsDvojQyaKuyC4ac_weSz4Xv0j0TuP8,14305
 empathy_os/trust_building.py,sha256=z2ZsQWK5QjX6d7pxFWrFD54J1keUguTog6YztZJVPKI,19180
-empathy_os/vscode_bridge 2.py,sha256=SgFmWFCoAoPww-g2hsRC2VeqAMTqUmXXxJ3hglykxP4,4967
 empathy_os/vscode_bridge.py,sha256=SgFmWFCoAoPww-g2hsRC2VeqAMTqUmXXxJ3hglykxP4,4967
 empathy_os/workflow_commands.py,sha256=tps6norBiw7Q2hV0qdqwdX4B7Kedt_FwYBQWhc2oEYg,25482
 empathy_os/adaptive/__init__.py,sha256=07m_Gsdxmm9mDBjJRFz7a8wZK-Tdd2QFRQOUw6yCc1w,342
@@ -137,23 +136,28 @@ empathy_os/cli/__init__.py,sha256=7PFURX1O4_sX14kWvNrnAV1Ndeixrx_nI8IN8eLQpQs,49
 empathy_os/cli/__main__.py,sha256=md0GsFpiTgifmJ0MJQ_E6Daq9JCxyCUgOUJRdJ69Ews,215
 empathy_os/cli/core.py,sha256=USK7MwpSqxKcMP0rxtf9qqTgQ1GMmcPpXBw9uadQ3sM,883
 empathy_os/cli/commands/__init__.py,sha256=GYdoNv80KS2p7bCThuoRbhiLmIZW6ddPqPEFEx567nQ,35
+empathy_os/cli/commands/batch.py,sha256=yyqKz_W9qchKlVFkCshKziFYZi6-1RCC0yp9j7tPCVk,7951
+empathy_os/cli/commands/cache.py,sha256=1KSX_4dv1ZfwtfGxhb5kop-8mSPBCkFFFAI9BCfh9Wk,8188
 empathy_os/cli/commands/help.py,sha256=AerAgAl0zW8YCCQIkxxc9p9K_wAskDGNhq211elPY_k,10082
 empathy_os/cli/commands/info.py,sha256=SFwOAhHjIoHRguqbMEPT0lG7bbt72Zv5AFpZjJPfHm8,4794
-empathy_os/cli/commands/inspect.py,sha256=W8RqFlErrjOYCUfM-o6uFXXDoeL9gEa9ZKj3D2qHFSY,16264
+empathy_os/cli/commands/inspect.py,sha256=DgqfA5uGEN94ROcoYja3Wa4KHCYPb0-df9N53pIUV24,16228
 empathy_os/cli/commands/inspection.py,sha256=1rszYfpj1zcYZsbKhJztQlwO1anF0Pgc5PGQ6M9Knk8,1781
 empathy_os/cli/commands/memory.py,sha256=1NTgfIHGdjKwsU5fAZZwZI5HUN7b4AE62_UF9psw3NE,1361
-empathy_os/cli/commands/metrics.py,sha256=TXFKVh00MYtTIAi_0YBeVsFCSgbZSpbNnI-mRCOMxVc,3178
+empathy_os/cli/commands/metrics.py,sha256=pX41tJpvRu-rGip2drdCkgZ0LDxOd0DVGXuV_hEY7ng,3176
 empathy_os/cli/commands/orchestrate.py,sha256=ifqEGUGaz8aF1mW3muGrbaJRqqJodcBXIyZDIIf1WPA,6553
 empathy_os/cli/commands/patterns.py,sha256=EVLX4akwGdkZOvJEQZDUojIAa5yK_50rE-TTFzEs-9w,7938
 empathy_os/cli/commands/profiling.py,sha256=n_1cyLwwguSHJUj6NKDr5gBuP1gSBLnkQ_ccxUmAkNY,5706
 empathy_os/cli/commands/provider.py,sha256=Mg8SjNsWKfhhdEZ0RvyfEaR3yiYcGj5L6OE5YFZRRFw,3095
+empathy_os/cli/commands/routing.py,sha256=2Z19yuWNeIbJeVDKrLzsNLkDn5m6gfYDA_TXbdVLJuo,9905
 empathy_os/cli/commands/setup.py,sha256=cF_3u7lrxfoyTr-ucu4D4Gc01EqtAQSxrwwU32tDsA0,3456
 empathy_os/cli/commands/status.py,sha256=ZIJx4Y0iMVFW_NDnoio1R6NBcXXS543QD9MwdaPA5B0,8309
 empathy_os/cli/commands/sync.py,sha256=5ABcmJ--puRXEH10yXBMOnqaue1miEPCzYhIbov5pq8,5094
 empathy_os/cli/commands/tier.py,sha256=3l3j7m83lcaKUFClTb2iU5vkH06zC8-eLGrt-KHdz0M,3745
 empathy_os/cli/commands/utilities.py,sha256=e4izGTS_uOy4YJDPwPbQnQ20p9M2v3JUyqAQEPwitMc,3737
-empathy_os/cli/commands/workflow.py,sha256=_SBhHGbjMtJaCgLN2XXhnYlcXlwGv4QCbSlQSas-7jE,23183
-empathy_os/cli/parsers/__init__.py,sha256=Z8sI0gvIH8n3Wu02TmtzSV80YYyQM65FAOoKqW7xBnI,1433
+empathy_os/cli/commands/workflow.py,sha256=UGmZcUVMDL2Dvg3umbsUYn48s0D73Cxy4XMIF3SgPzQ,23183
+empathy_os/cli/parsers/__init__.py,sha256=Oihx1jb5wN_4GS3RxZiEYhQhglI44idOOlFJs9a6e5g,1677
+empathy_os/cli/parsers/batch.py,sha256=wSutRqerJZxUNVNIotU9qtI8EGP6IzVihEZUViCItCQ,3236
+empathy_os/cli/parsers/cache.py,sha256=_SpFE48JM1xfAvb_ndqYNR0bP4EyMb_T_PlmIsIHf5o,1797
 empathy_os/cli/parsers/help.py,sha256=I3QFDVABxZpYDCJ52GNWwfQlKEdV72Yr9A8spPHqUqU,1642
 empathy_os/cli/parsers/info.py,sha256=525oPlxKkOJk8tra_a2ga9DKm6VtejQUBWzBCDQK1v4,989
 empathy_os/cli/parsers/inspect.py,sha256=DekJWgmrD_rVjUxvS0WI_Lx55DqquXZn_T0J3mC53M0,2427
@@ -161,6 +165,7 @@ empathy_os/cli/parsers/metrics.py,sha256=dbMIbmsrGxRvaG6kMh8LB3IWtmD_R4u6W131skV
 empathy_os/cli/parsers/orchestrate.py,sha256=kACmZExVIJ_F1cBRGOYtSxf8AWy_k1n3GNTuDIsxNVY,1769
 empathy_os/cli/parsers/patterns.py,sha256=4Jy7A_iZWJRnIdPUi6vz2tl2_agkhSgpNv8LlEoUScI,2374
 empathy_os/cli/parsers/provider.py,sha256=fh2YGIQ0LOP7Lb2LaphOQFO01rWDxhvCprXYWLy7YXE,1115
+empathy_os/cli/parsers/routing.py,sha256=YJ4W_UISXCOpvb7xp4GZUbJhBpKDg5ew7izJSzq8B6E,2884
 empathy_os/cli/parsers/setup.py,sha256=K3bghLEkNijDAAvmUfBnDFOsG9J9Hpmyi7EruFalQaY,1105
 empathy_os/cli/parsers/status.py,sha256=Z_r-C92hcrJ3iBmSok6iEkZfYDRvs7_3WD0oQQULzDw,2370
 empathy_os/cli/parsers/sync.py,sha256=fYXJDJTE1_JYdc15y7bHJOalCnB0Yg9ZkQQPPTUMUIo,884
@@ -174,7 +179,7 @@ empathy_os/config/xml_config.py,sha256=iXFJvGPlSoxRdARlxx-7YIdxb8eRiFHVRWM6-HwOW
 empathy_os/dashboard/__init__.py,sha256=CwGq_IQS5H3OjY0NabAihvxohns8vsw0jmEnq3p1UGg,1413
 empathy_os/dashboard/app.py,sha256=taCHkhSAsnna_joClhLTai5ayevYbh67nJDpOYOU-EA,16147
 empathy_os/dashboard/simple_server.py,sha256=L2uSgCJtiHS7TZSu-iEXpImNTAW4w3zNVg8YNsj5vPs,14667
-empathy_os/dashboard/standalone_server.py,sha256=c6csNtVFTeQNgaCTYC7EhC9A5AyQab8wbFPHy07rZdE,20331
+empathy_os/dashboard/standalone_server.py,sha256=rsJdlHZii3UUbA4Yi3H8KDmc77RGqBbyZUHIOKvFE2M,20862
 empathy_os/hot_reload/README.md,sha256=FnQWSX-dfTIJvXHjjiWRWBjMK2DR-eyOfDHJlGIzm0o,10406
 empathy_os/hot_reload/__init__.py,sha256=Aos2tLSKRzOLr5zRomLyzrz6qDRd1_PlinI3vrJcCTo,1642
 empathy_os/hot_reload/config.py,sha256=Lk_5bShouV-Z_atOvgcoPas4wwXwfHTKK9eSC60C9Uk,2290
@@ -215,6 +220,7 @@ empathy_os/meta_workflows/session_context.py,sha256=MNx_P7Sv-Tt5YXaX-HH2HUc_f2sc
 empathy_os/meta_workflows/template_registry.py,sha256=jEvQNZtQ4UhitBuyml1dDjQJNLoT-BvPIFgvWSuw0xs,7485
 empathy_os/meta_workflows/workflow.py,sha256=cKzaIWcVuRMRk2YoXJCMnl_4y5XUpD6b-uVStAlUqe8,36007
 empathy_os/metrics/__init__.py,sha256=b0lkly5Fz89AvvGqV6lLYx8hWlmttL1ZCPuWlHjwXxY,369
+empathy_os/metrics/collector.py,sha256=WBrUklvZ9T80a5iz1XyRPvmteMOpftqBYHdBb4DpZao,766
 empathy_os/metrics/prompt_metrics.py,sha256=C9N8uIynTP8XiGcwUeOoJRU48vLiS3uK0SKFmVefalo,6518
 empathy_os/models/__init__.py,sha256=BixzbjDbweDr7siPxAKGvoFdYY5M3hxoml7pIiVSpPE,3485
 empathy_os/models/__main__.py,sha256=V7uZJ_XT__ROZsmnEI4-U7hNsOH1oqzfnWb2o6xJ0g4,296
@@ -227,7 +233,7 @@ empathy_os/models/provider_config.py,sha256=RMGjWqjOCSUGEtudQU7uaTHQj1RgZFR9BlKd
 empathy_os/models/registry.py,sha256=L9GZ5hQvCcGDcaorOTW7GX4Ndu1wnqxXtwSjQ5cT0_4,15555
 empathy_os/models/tasks.py,sha256=mvxFZiiRgUGCHYdJE316Zp8PGhN5D1Ctx9Og3jz_Zy8,10396
 empathy_os/models/telemetry.py,sha256=piNLJXmLvpWQQu6gFHcUpaLEaPmFzvzKTmO1TGEbfOU,52532
-empathy_os/models/token_estimator.py,sha256=y242dDja7oZSg0LsfxYBmsGAaoiyVQ-bNWTQXV4CI_I,13042
+empathy_os/models/token_estimator.py,sha256=15RsTew2aqA6yNuaUFDLX_CMjkIjRELWeWvlkDV0G4g,13488
 empathy_os/models/validation.py,sha256=JYOXAy9Bj50uO9LHDbBRBK73sue8TLVVRtp6vhzp7ms,9152
 empathy_os/monitoring/__init__.py,sha256=efAzUX0DYeUTGEES-MV6jI4iUYmhH-RkQYnVLnjSl-M,1507
 empathy_os/monitoring/alerts.py,sha256=NZUbBXC_a5TD235cs3G-XKYxLIBpVCENyTW3zXjjUss,30281
@@ -297,12 +303,12 @@ empathy_os/socratic/success.py,sha256=wZoHBkfkzmZwwbEMNdUj8cWw4nLCyKFtGNcWHJDoV9
 empathy_os/socratic/visual_editor.py,sha256=Nk2vJaTKtcwhHH04KTzRnAVyae6AeA03je2-y74d1kk,27912
 empathy_os/socratic/web_ui.py,sha256=Sg7pSS1043ecVt_yYpSLnv_6pvGnt6aM742fXq6uJ7M,25375
 empathy_os/telemetry/__init__.py,sha256=DpNi4Eglyj7mAA4XT3GFpidnvvbVNC7Qo1Te9Q5wQ44,1295
-empathy_os/telemetry/agent_coordination.py,sha256=ZqtXFq-GR5vdNO18taJYwW78jdzXeWJBMumxiPvTvHo,16072
-empathy_os/telemetry/agent_tracking.py,sha256=o49YUXcVnwfMBFUDKPh2Cw2FfbhL5oIFhk5HecKHVNQ,11655
-empathy_os/telemetry/approval_gates.py,sha256=WwYolC-3LKR8ippu5qna1GUtiAmZfuHFwMt9vcnxyvo,19933
+empathy_os/telemetry/agent_coordination.py,sha256=GhiPcSPzgQ032b6r1c86Ktf_2TcH63e23wkxLMm_fME,15898
+empathy_os/telemetry/agent_tracking.py,sha256=cA_vqLSmmqVkeaiZLHLn38BoGeB1Y8VE7mRIU1dWClU,11672
+empathy_os/telemetry/approval_gates.py,sha256=1h3vZyD3rBdVQ2og3kTsXtDFm52sf6w5j1XU4lPEYoM,19300
 empathy_os/telemetry/cli.py,sha256=vuWuAgq8VyZBogqnD0DHI76MCg8WLVsjyTDKM1uh1sY,69474
-empathy_os/telemetry/event_streaming.py,sha256=-tMNmI3eeDTDUMoE13nH9Gm6_3JVAft0Ic2kk7zl8qU,13096
-empathy_os/telemetry/feedback_loop.py,sha256=PLfR4EJuXlyVhJMxKcgNwXrLy7nP8cxiLX_21WRI4Aw,19423
+empathy_os/telemetry/event_streaming.py,sha256=HSLBTw2svys-JKV530ht1CcFP9Ix038qHwbS424vY4g,13223
+empathy_os/telemetry/feedback_loop.py,sha256=nKrm7tYKYXbmLZew2M4cETRhDbv7qerK8WZMsULhYx8,19314
 empathy_os/telemetry/usage_tracker.py,sha256=Be409JSeweps3GEN3k66u_0I1594KTWoH_bRzelih9U,21184
 empathy_os/test_generator/__init__.py,sha256=lSck9qlC32AO8qoQldk3UjjTRDPdAUgIOHGa-WvzCqI,919
 empathy_os/test_generator/__main__.py,sha256=YY_HE1xg4zKZkHHAd6sSzWvJCvLFOtmpawCrNlGjWAc,345
@@ -321,7 +327,7 @@ empathy_os/workflow_patterns/registry.py,sha256=0U_XT0hdQ5fLHuEJlrvzjaCBUyeWDA67
 empathy_os/workflow_patterns/structural.py,sha256=v1wbBU0pCQQKdNhPXM_fuPO8ptQCbEC_8X-OKgfhrk8,9432
 empathy_os/workflows/__init__.py,sha256=iaOjOtiJzfLBKey5W_v1LT6o-yjrsWLbQNglH8tDN5I,19594
 empathy_os/workflows/base.py,sha256=5pKLwkg6qULFhT2QFY5nMuxUnYLq__ZGL5ogNQcmDNA,99363
-empathy_os/workflows/batch_processing.py,sha256=1DO5IQV7_LKTm0Mkcx0wZqrT_MFMQF-MKdxep64TppQ,9645
+empathy_os/workflows/batch_processing.py,sha256=hMODlPHV74iorBz9_405LqKYhk-3twIgmUyMRwATqGQ,11600
 empathy_os/workflows/bug_predict.py,sha256=x49KCzdsa6t1Em0FBpFCigOj0FB-zhydW-Pc2jTVeYE,38378
 empathy_os/workflows/builder.py,sha256=rRXevtS8v1FAGG91yhpSq6_dx5rvK_sZrF9ekdW40fE,8655
 empathy_os/workflows/caching.py,sha256=y6Ws0KKCQ_R6D6PTbpc8dQRjb3tG4U2-KcwJvxFi07g,8134
@@ -375,23 +381,14 @@ empathy_os/workflows/keyboard_shortcuts/parsers.py,sha256=aws4HSjqBOrl-DQEOV9WeJ
 empathy_os/workflows/keyboard_shortcuts/prompts.py,sha256=gcV2F2bAMjZUrbB13lOI4ixXzXm2TNWEZ4VbPhC7ITw,9164
 empathy_os/workflows/keyboard_shortcuts/schema.py,sha256=MwvM63J9WTO6nqtwes5A04HH1dTa9XhJlD0SbFhsS5E,5806
 empathy_os/workflows/keyboard_shortcuts/workflow.py,sha256=FOEQwc8IR0PEYim05zjR-uIlw6oPZHW9jQZ8skxXp0c,17678
-empathy_os/workflows/progressive/README 2.md,sha256=ngn3ZC48LW-ON5Vow90dW4UvEWeOKuU7V-W4JdudqJo,13793
 empathy_os/workflows/progressive/README.md,sha256=ngn3ZC48LW-ON5Vow90dW4UvEWeOKuU7V-W4JdudqJo,13793
-empathy_os/workflows/progressive/__init__ 2.py,sha256=l18JX30ONk8H08Wm210k1W3AKat2tVzI840JqlzTBpU,2203
 empathy_os/workflows/progressive/__init__.py,sha256=qkPVjO_MtHjxVcHMr5jbU9NMZTxVCGHXVD3EkngyCso,2179
-empathy_os/workflows/progressive/cli 2.py,sha256=WIFxTBUsCL2hvKKoPFxYUgfhRFjMm-HVUjRp3Y-pvXA,6305
 empathy_os/workflows/progressive/cli.py,sha256=jbl5jSk3wlbjGc00Ao06OO7aU53AZuVZVKJpJJAajX4,6122
-empathy_os/workflows/progressive/core 2.py,sha256=891OttvWkFokBRVzvqVUxwpcNlgc7edlVYqGUCkn0vs,16101
 empathy_os/workflows/progressive/core.py,sha256=2FzIsy0xKyP91-aMTGfERbX5Mow5CSlfR9RUmHU4GeM,16070
-empathy_os/workflows/progressive/orchestrator 2.py,sha256=FsLcoRC8jujRcOV87PyN-0jzlLvStIQQqMFisxWWg0I,26786
 empathy_os/workflows/progressive/orchestrator.py,sha256=LssTqdy4ufqhgpgS3QNZEdpeSkdg-MgpFyhW5zCJmAY,27733
-empathy_os/workflows/progressive/reports 2.py,sha256=F2p3HyInGH8fT9-e_BsOVcgeQBtOugYsSyBn49yhhmc,17588
 empathy_os/workflows/progressive/reports.py,sha256=VgymV6v7owodELf_QLQTwqGOEZjaFKclKxeQ0-vow0k,17631
-empathy_os/workflows/progressive/telemetry 2.py,sha256=5ZjCmeGjtfESejocWdtvnm5iCTAQv5_3NRTYmuoIbVs,9427
 empathy_os/workflows/progressive/telemetry.py,sha256=DiDTrkeKDjsWJnA57UrNfhbSG6qv9LaTP2wqHPcRscA,9398
-empathy_os/workflows/progressive/test_gen 2.py,sha256=rNjjrQ8AFoXRe99X1Yk020gX6tcTxLh5pGMXXho8XRs,15802
 empathy_os/workflows/progressive/test_gen.py,sha256=wdpjx-Tlz-dWqOwHwoblbgEABr7EjTRV2faXMTad1M0,15759
-empathy_os/workflows/progressive/workflow 2.py,sha256=bGXKmB9UFLfDw5c9MsW6Zw4GSbknCP-9HnaIEX36nhY,21597
 empathy_os/workflows/progressive/workflow.py,sha256=FD3MzRSMV1n_c1S8jSlLLIrcewcus6Niyml6-EMdk_E,21146
 empathy_software_plugin/SOFTWARE_PLUGIN_README.md,sha256=XjvK2TXoI5nsgEZqh4RQIAxBiGVpvbXihIu7zF8KNYk,1399
 empathy_software_plugin/__init__.py,sha256=OceDhLoQrjD0jTPlI4418608l97OT48u4SCL5yEDPQc,309
@@ -403,8 +400,8 @@ workflow_scaffolding/__init__.py,sha256=UpX5vjjjPjIaAKyIV1D4GxJzLUZy5DzdzgSkePYM
 workflow_scaffolding/__main__.py,sha256=0qspuNoadTDqyskXTlT8Sahqau-XIxN35NHTSGVW6z4,236
 workflow_scaffolding/cli.py,sha256=yLgvMsPbqI-LHt7UKXMuj-Dc0-44EqWiChldysrw1YQ,6763
 workflow_scaffolding/generator.py,sha256=9eEmm324kVzf8BCtXEML090pyke06o0U7aGATzZIaPM,8869
-empathy_framework-5.0.1.dist-info/METADATA,sha256=V8SL6cr3yxanS5sBbVaSn6ywChFbsqEE9tK7D0fHBWY,31268
-empathy_framework-5.0.1.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
-empathy_framework-5.0.1.dist-info/entry_points.txt,sha256=Ox9iu_2Fx66x5fyZ5L1NJrrFFekD-BZJIudHNzKwYPs,1565
-empathy_framework-5.0.1.dist-info/top_level.txt,sha256=YJNBYFE2u5f2XnveFqYO-rZDcHR_aMOHAiIIf9tl72c,102
-empathy_framework-5.0.1.dist-info/RECORD,,
+empathy_framework-5.0.3.dist-info/METADATA,sha256=wyKjpZD6hL70hezsVORRmmaRUR3gkY8TjTl5xaJ0Psk,32879
+empathy_framework-5.0.3.dist-info/WHEEL,sha256=wUyA8OaulRlbfwMtmQsvNngGrxQHAvkKcvRmdizlJi0,92
+empathy_framework-5.0.3.dist-info/entry_points.txt,sha256=Ox9iu_2Fx66x5fyZ5L1NJrrFFekD-BZJIudHNzKwYPs,1565
+empathy_framework-5.0.3.dist-info/top_level.txt,sha256=YJNBYFE2u5f2XnveFqYO-rZDcHR_aMOHAiIIf9tl72c,102
+empathy_framework-5.0.3.dist-info/RECORD,,

empathy_llm_toolkit/providers.py

@@ -322,6 +322,93 @@ class AnthropicProvider(BaseLLMProvider):
             },
         )
 
+    def estimate_tokens(self, text: str) -> int:
+        """Estimate token count using accurate token counter (overrides base class).
+
+        Uses tiktoken for fast local estimation (~98% accurate).
+        Falls back to heuristic if tiktoken unavailable.
+
+        Args:
+            text: Text to count tokens for
+
+        Returns:
+            Estimated token count
+        """
+        try:
+            from .utils.tokens import count_tokens
+
+            return count_tokens(text, model=self.model, use_api=False)
+        except ImportError:
+            # Fallback to base class heuristic if utils not available
+            return super().estimate_tokens(text)
+
+    def calculate_actual_cost(
+        self,
+        input_tokens: int,
+        output_tokens: int,
+        cache_creation_tokens: int = 0,
+        cache_read_tokens: int = 0,
+    ) -> dict[str, Any]:
+        """Calculate actual cost based on precise token counts.
+
+        Includes Anthropic prompt caching cost adjustments:
+        - Cache writes: 25% markup over standard input pricing
+        - Cache reads: 90% discount from standard input pricing
+
+        Args:
+            input_tokens: Regular input tokens (not cached)
+            output_tokens: Output tokens
+            cache_creation_tokens: Tokens written to cache
+            cache_read_tokens: Tokens read from cache
+
+        Returns:
+            Dictionary with cost breakdown:
+            - base_cost: Cost for regular input/output tokens
+            - cache_write_cost: Cost for cache creation (if any)
+            - cache_read_cost: Cost for cache reads (if any)
+            - total_cost: Total cost including all components
+            - savings: Amount saved by cache reads vs. full price
+
+        Example:
+            >>> provider = AnthropicProvider(api_key="...")
+            >>> cost = provider.calculate_actual_cost(
+            ...     input_tokens=1000,
+            ...     output_tokens=500,
+            ...     cache_read_tokens=10000
+            ... )
+            >>> cost["total_cost"]
+            0.0105  # Significantly less than without cache
+        """
+        # Get pricing for this model
+        model_info = self.get_model_info()
+        input_price_per_million = model_info["cost_per_1m_input"]
+        output_price_per_million = model_info["cost_per_1m_output"]
+
+        # Base cost (non-cached tokens)
+        base_cost = (input_tokens / 1_000_000) * input_price_per_million
+        base_cost += (output_tokens / 1_000_000) * output_price_per_million
+
+        # Cache write cost (25% markup)
+        cache_write_price = input_price_per_million * 1.25
+        cache_write_cost = (cache_creation_tokens / 1_000_000) * cache_write_price
+
+        # Cache read cost (90% discount = 10% of input price)
+        cache_read_price = input_price_per_million * 0.1
+        cache_read_cost = (cache_read_tokens / 1_000_000) * cache_read_price
+
+        # Calculate savings from cache reads
+        full_price_for_cached = (cache_read_tokens / 1_000_000) * input_price_per_million
+        savings = full_price_for_cached - cache_read_cost
+
+        return {
+            "base_cost": round(base_cost, 6),
+            "cache_write_cost": round(cache_write_cost, 6),
+            "cache_read_cost": round(cache_read_cost, 6),
+            "total_cost": round(base_cost + cache_write_cost + cache_read_cost, 6),
+            "savings": round(savings, 6),
+            "currency": "USD",
+        }
+
 
 class AnthropicBatchProvider:
     """Provider for Anthropic Batch API (50% cost reduction).
@@ -370,7 +457,8 @@ class AnthropicBatchProvider:
         """Create a batch job.
 
         Args:
-            requests: List of request dicts with 'custom_id', 'model', 'messages', etc.
+            requests: List of request dicts with 'custom_id' and 'params' containing message creation parameters.
+                Format: [{"custom_id": "id1", "params": {"model": "...", "messages": [...], "max_tokens": 1024}}]
             job_id: Optional job identifier for tracking (unused, for API compatibility)
 
         Returns:
@@ -384,22 +472,46 @@ class AnthropicBatchProvider:
             >>> requests = [
             ...     {
             ...         "custom_id": "task_1",
-            ...         "model": "claude-sonnet-4-5",
-            ...         "messages": [{"role": "user", "content": "Test"}],
-            ...         "max_tokens": 1024
+            ...         "params": {
+            ...             "model": "claude-sonnet-4-5-20250929",
+            ...             "messages": [{"role": "user", "content": "Test"}],
+            ...             "max_tokens": 1024
+            ...         }
             ...     }
             ... ]
             >>> batch_id = provider.create_batch(requests)
             >>> print(f"Batch created: {batch_id}")
-            Batch created: batch_abc123
+            Batch created: msgbatch_abc123
         """
         if not requests:
             raise ValueError("requests cannot be empty")
 
+        # Validate and convert old format to new format if needed
+        formatted_requests = []
+        for req in requests:
+            if "params" not in req:
+                # Old format: convert to new format with params wrapper
+                formatted_req = {
+                    "custom_id": req.get("custom_id", f"req_{id(req)}"),
+                    "params": {
+                        "model": req.get("model", "claude-sonnet-4-5-20250929"),
+                        "messages": req.get("messages", []),
+                        "max_tokens": req.get("max_tokens", 4096),
+                    },
+                }
+                # Copy other optional params
+                for key in ["temperature", "system", "stop_sequences"]:
+                    if key in req:
+                        formatted_req["params"][key] = req[key]
+                formatted_requests.append(formatted_req)
+            else:
+                formatted_requests.append(req)
+
         try:
-            batch = self.client.batches.create(requests=requests)
+            # Use correct Message Batches API endpoint
+            batch = self.client.messages.batches.create(requests=formatted_requests)
             self._batch_jobs[batch.id] = batch
-            logger.info(f"Created batch {batch.id} with {len(requests)} requests")
+            logger.info(f"Created batch {batch.id} with {len(formatted_requests)} requests")
             return batch.id
         except Exception as e:
             logger.error(f"Failed to create batch: {e}")
@@ -412,18 +524,20 @@ class AnthropicBatchProvider:
             batch_id: Batch job ID
 
         Returns:
-            BatchStatus object with status field:
-            - "processing": Batch is being processed
-            - "completed": Batch processing completed
-            - "failed": Batch processing failed
+            MessageBatch object with processing_status field:
+            - "in_progress": Batch is being processed
+            - "canceling": Cancellation initiated
+            - "ended": Batch processing ended (check request_counts for success/errors)
 
         Example:
-            >>> status = provider.get_batch_status("batch_abc123")
-            >>> print(status.status)
-            processing
+            >>> status = provider.get_batch_status("msgbatch_abc123")
+            >>> print(status.processing_status)
+            in_progress
+            >>> print(f"Succeeded: {status.request_counts.succeeded}")
         """
         try:
-            batch = self.client.batches.retrieve(batch_id)
+            # Use correct Message Batches API endpoint
+            batch = self.client.messages.batches.retrieve(batch_id)
             self._batch_jobs[batch_id] = batch
             return batch
         except Exception as e:
@@ -437,25 +551,37 @@ class AnthropicBatchProvider:
             batch_id: Batch job ID
 
         Returns:
-            List of result dicts matching input order
+            List of result dicts. Each dict contains:
+            - custom_id: Request identifier
+            - result: Either {"type": "succeeded", "message": {...}} or {"type": "errored", "error": {...}}
 
         Raises:
-            ValueError: If batch is not completed
+            ValueError: If batch has not ended processing
             RuntimeError: If API call fails
 
         Example:
-            >>> results = provider.get_batch_results("batch_abc123")
+            >>> results = provider.get_batch_results("msgbatch_abc123")
             >>> for result in results:
-            ...     print(f"{result['custom_id']}: {result['response']['content']}")
+            ...     if result['result']['type'] == 'succeeded':
+            ...         message = result['result']['message']
+            ...         print(f"{result['custom_id']}: {message.content[0].text}")
+            ...     else:
+            ...         error = result['result']['error']
+            ...         print(f"{result['custom_id']}: Error {error['type']}")
         """
         status = self.get_batch_status(batch_id)
 
-        if status.status != "completed":
-            raise ValueError(f"Batch {batch_id} not completed (status: {status.status})")
+        # Check processing_status instead of status
+        if status.processing_status != "ended":
+            raise ValueError(
+                f"Batch {batch_id} has not ended processing (status: {status.processing_status})"
+            )
 
         try:
-            results = self.client.batches.results(batch_id)
-            return list(results)
+            # Use correct Message Batches API endpoint
+            # results() returns an iterator, convert to list
+            results_iterator = self.client.messages.batches.results(batch_id)
+            return list(results_iterator)
         except Exception as e:
             logger.error(f"Failed to get batch results for {batch_id}: {e}")
             raise RuntimeError(f"Failed to get batch results: {e}") from e
@@ -474,15 +600,15 @@ class AnthropicBatchProvider:
             timeout: Maximum wait time in seconds (default: 86400 = 24 hours)
 
         Returns:
-            Batch results when completed
+            Batch results when processing ends
 
         Raises:
             TimeoutError: If batch doesn't complete within timeout
-            RuntimeError: If batch processing fails
+            RuntimeError: If batch had errors during processing
 
         Example:
             >>> results = await provider.wait_for_batch(
-            ...     "batch_abc123",
+            ...     "msgbatch_abc123",
             ...     poll_interval=300,  # Check every 5 minutes
             ... )
             >>> print(f"Batch completed: {len(results)} results")
@@ -493,22 +619,36 @@ class AnthropicBatchProvider:
         while True:
             status = self.get_batch_status(batch_id)
 
-            if status.status == "completed":
-                logger.info(f"Batch {batch_id} completed successfully")
-                return self.get_batch_results(batch_id)
+            # Check if batch processing has ended
+            if status.processing_status == "ended":
+                # Check request counts to see if there were errors
+                counts = status.request_counts
+                logger.info(
+                    f"Batch {batch_id} ended: "
+                    f"{counts.succeeded} succeeded, {counts.errored} errored, "
+                    f"{counts.canceled} canceled, {counts.expired} expired"
+                )
 
-            if status.status == "failed":
-                error_msg = getattr(status, "error", "Unknown error")
-                logger.error(f"Batch {batch_id} failed: {error_msg}")
-                raise RuntimeError(f"Batch {batch_id} failed: {error_msg}")
+                # Return results even if some requests failed
+                # The caller can inspect individual results for errors
+                return self.get_batch_results(batch_id)
 
             # Check timeout
             elapsed = (datetime.now() - start_time).total_seconds()
             if elapsed > timeout:
                 raise TimeoutError(f"Batch {batch_id} did not complete within {timeout}s")
 
-            # Log progress
-            logger.debug(f"Batch {batch_id} status: {status.status} (elapsed: {elapsed:.0f}s)")
+            # Log progress with request counts
+            try:
+                counts = status.request_counts
+                logger.debug(
+                    f"Batch {batch_id} status: {status.processing_status} "
+                    f"(processing: {counts.processing}, elapsed: {elapsed:.0f}s)"
+                )
+            except AttributeError:
+                logger.debug(
+                    f"Batch {batch_id} status: {status.processing_status} (elapsed: {elapsed:.0f}s)"
+                )
 
             # Wait before next poll
             await asyncio.sleep(poll_interval)