short id will generate based on full id when required

…d delta fitness

…regression feature weights

… and improve docstring clarity

refactor: memory system modular refactor + adversarial hardening

…cstrings

Phase 1 cleanup (completed):
- Move A_mem/, GAM_root/ → _vendor/ (vendored MIT libs)
- Move contrib licenses → _vendor/
- Move 3 example scripts → examples/
- Fix 15 broken vendored library imports (A_mem/GAM_root bare imports)
- Update 8 consumer import paths to _vendor/
- Add _vendor/__init__.py docstring (vendored libs notice)
- Add examples/__init__.py docstring (not production code)
- Update shared_memory/__init__.py docstring
- Update pyproject.toml (ruff/mypy exclude paths for vendors)

Tests: 770 passed
Lint: clean

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

refactor(memory): Phase 1 — directory reorg (vendor/examples/docstrings)

- Delete 5 duplicated usage-merge functions from memory_write_example.py
  (_to_float, _median_or_none, _extract_usage_task_deltas,
  _build_usage_payload_from_task_deltas, _merge_usage_payloads)
  → import from card_update_dedup.py (canonical home)
- Delete duplicate dedupe_keep_order from card_update_dedup.py
  → import from shared_memory/utils.py
- Remove deprecated _apply_update_actions() from memory.py (dead wrapper)
- Make memory_to_card private (_memory_to_card) — only used internally
- Simplify single-iteration loop in _extract_json_object

Net: ~120 lines deleted, zero behavior change.
Tests: 770 passed | Lint: clean

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

refactor(memory): deduplicate code and delete dead paths

Replace hand-written RetrievalWeights.from_mapping() and
CardUpdateDedupConfig.from_mapping() dict parsers (~87 lines) with
Pydantic v2 @model_validator(mode="before") — same behavior, idiomatic.

Also: add docstrings to all functions in card_update_dedup.py, fix
stale test references to deleted _apply_update_actions wrapper, add
test for flat config format.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

refactor(memory): Pydantic-idiomatic config parsing

- memory_write_example.py → write_pipeline.py (it's production, not an example)
- memory_write_config.py → write_pipeline_config.py
- selected_ideas_6.py → origin_analysis.py (remove versioned filename)
- Delete test_memory_write_example_extended.py (duplicate of test_write_pipeline.py)
- Update all 8 import sites + 1 dynamic importlib.import_module call

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

refactor(memory): rename write pipeline and analysis files

Extract from card_conversion.py (554 → 420 lines):
- base.py: GigaEvoMemoryBase abstract class (20 lines)
- card_search.py: format_search_results, search_cards_by_keyword,
  synthesize_search_results (115 lines)

Update 4 import sites directly (no re-exports).
card_conversion.py retains: normalization, conversion, GAM config,
constants, protocols.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

refactor(memory): split card_conversion into focused modules

Define MemoryError, MemoryRetrieverError, MemorySearchError, and
MemoryStorageError in gigaevo/exceptions.py following the existing
GigaEvoError hierarchy.

Wire them into the memory subsystem:
- gam_search.build() wraps all failures in MemoryRetrieverError
- memory.py narrows two gam.build() catches from bare Exception
- card_store._load() narrows to (json.JSONDecodeError, OSError)
- card_dedup import block narrows to (ImportError, OSError)

Resilience-critical catches (search fallback, merge loop, __exit__)
remain broad by design.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

refactor(memory): custom exception hierarchy and narrowed catches

…t base to ABC

- concept_api.py: all 5 RuntimeError raises → MemoryStorageError
  (matches gigaevo/database pattern of wrapping I/O errors)
- base.py: GigaEvoMemoryBase now uses ABC + @AbstractMethod
  (matches MutationOperator, Stage, LangGraphAgent pattern)
- card_dedup.py: narrow two broad catches:
  - JSONL read fallback: except Exception → (json.JSONDecodeError, OSError)
  - GAM store build: except Exception → (MemoryRetrieverError, OSError)
- Update 6 test assertions from RuntimeError to MemoryStorageError

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

…ormity

refactor(memory): exception conformity + ABC base class

When write_pipeline.py passes MemoryCard/ProgramCard Pydantic models to
memory_platform.save_card(), the dict() call on a Pydantic model doesn't
properly flatten nested Pydantic objects like ConnectedIdea. This caused
TypeError in _persist_index() when json.dumps() tried to serialize.

Root cause: write_pipeline returns list[AnyCard] (Pydantic models) and both
backends (memory_platform and memory/shared_memory) consume these cards via
save_card(). memory_platform's normalize_memory_card() must explicitly call
.model_dump() on Pydantic inputs to flatten nested objects.

Fix verified: all 788 memory + integration tests pass.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Tests the exact bug path: Pydantic MemoryCard/ProgramCard with nested
ConnectedIdea and MemoryCardExplanation objects must be properly
flattened to plain dicts before JSON serialization.

6 tests covering: ProgramCard with ConnectedIdea, MemoryCard with
MemoryCardExplanation, plain dict passthrough, JSON round-trips, None.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

- Add gigaevo-memory Git dependency to pyproject.toml
- Remove sys.path manipulation from memory_platform/memory.py and
  remote_gam_retriever.py (no longer needed with proper install)
- Simplify test file to use direct imports instead of module mocking

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Expands from 6 to 11 tests covering the complete save_card → _persist_index
flow with Pydantic inputs. Tests verify:

- normalize_memory_card: ConnectedIdea/MemoryCardExplanation → dict
- save_card: Pydantic ProgramCard/MemoryCard → JSON-serializable index
- _card_to_backend_content: API payload is clean dict
- persist/reload roundtrip: index file survives write→read cycle

Uses _make_platform_memory() factory with mocked API client to test
memory_platform in isolation without network dependencies.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

- Add docstrings to 15 public methods across 5 files (memory.py,
  concept_api.py, card_dedup.py, openai_inference.py, write_pipeline.py)
- Add return type annotations to 4 functions in amem_gam_retriever.py
- Fix 2 mypy errors: annotate retrievers dict, rename variable in api_sync.py
- Extract magic numbers: _MAX_SUMMARY_CHARS, _MAX_DESCRIPTION_CHARS,
  _ENTITY_NAME_MAX_LENGTH, _MAX_CONNECTED_DESCRIPTIONS

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

refactor(memory): type annotations, docstrings, constants, platform bug fix

on_mutation_outcome forwarded program.get_metadata('mutation_model')
directly to self._bandit.update_reward, which raised KeyError when
the metadata did not match a current arm (Redis snapshot replay,
custom operator, hand-built test fixture). Membership check at the
top of the function turns the crash into a debug-level skip.