fix: universal parsing + OpenRouter + state bug — purpose_agent/robust_parser.py

fix: real-model robustness — benchmarks/validate_real.py

fix: real-model robustness — purpose_agent/purpose_function.py

fix: real-model robustness — purpose_agent/actor.py

Track 2: validation suite with improvement curves, cold/warm, transfer, adversarial

Track 2: validation suite with improvement curves, cold/warm, transfer, adversarial

Track 2: validation suite with improvement curves, cold/warm, transfer, adversarial

Track 1: tests/test_core.py

Track 1: examples/quickstart.py

Track 1: pyproject.toml

Track 1: purpose_agent/slm_backends.py

Track 1: purpose_agent/experience_replay.py

Track 1: purpose_agent/tools.py

Track 1: purpose_agent/llm_backend.py

Track 1: purpose_agent/__init__.py

release: Purpose Agent v2.0.0 — final README with 13-paper architecture

release: Purpose Agent v2.0.0 — updated __init__.py with all V2 + research exports

V2 merge: COMPILED_RESEARCH.md

V2 merge: purpose_agent/unified.py

V2 merge: purpose_agent/tools.py

V2 merge: purpose_agent/retroformer.py

V2 merge: purpose_agent/llm_compiler.py

V2 merge: purpose_agent/prompt_optimizer.py

V2 merge: purpose_agent/self_taught.py

V2 merge: purpose_agent/meta_rewarding.py

V2 merge: purpose_agent/benchmark_v2.py

V2 merge: purpose_agent/evalport.py

V2 merge: purpose_agent/memory_ci.py

V2 merge: purpose_agent/immune.py

V2 merge: purpose_agent/compiler.py

V2 merge: purpose_agent/memory.py

V2 merge: purpose_agent/trace.py

V2 merge: purpose_agent/v2_types.py

feat: Rewrite README — lead with easy API, 3 usage levels, complete docs

feat: Easy API — purpose(), Team, quickstart wizard, template auto-detection

feat: Easy API — purpose(), Team, quickstart wizard, template auto-detection

feat: Easy API — purpose(), Team, quickstart wizard, template auto-detection

docs: add unified capabilities research to COMPILED_RESEARCH.md

feat: unified capabilities (Graph + Parallel + Conversation + Agent + KnowledgeStore)

feat: unified capabilities (Graph + Parallel + Conversation + Agent + KnowledgeStore)

feat: unified capabilities (Graph + Parallel + Conversation + Agent + KnowledgeStore)

refactor: modularity fixes + plugin registry + compiled research

refactor: modularity fixes + plugin registry + compiled research

refactor: modularity fixes + plugin registry + compiled research

refactor: modularity fixes + plugin registry + compiled research

refactor: modularity fixes + plugin registry + compiled research

refactor: modularity fixes + plugin registry + compiled research

refactor: modularity fixes + plugin registry + compiled research

v0.2.0: Complete README with SLM-native, multi-agent, HITL, eval, streaming, tools, observability

v0.2.0: Add purpose_agent/evaluation.py