getzep · mehulp93 · Aug 7, 2026 · Aug 1, 2026 · Aug 1, 2026 · Aug 1, 2026
diff --git a/.github/workflows/test-integrations.yml b/.github/workflows/test-integrations.yml
@@ -42,6 +42,8 @@ jobs:
               - 'integrations/langgraph/python/**'
             pydantic-ai:
               - 'integrations/pydantic-ai/python/**'
+            strands:
+              - 'integrations/strands/python/**'
 
       - uses: dorny/paths-filter@d1c1ffe0248fe513906c8e24db8ea791d46f8590 # v3
         id: typescript

diff --git a/README.md b/README.md
@@ -49,7 +49,7 @@ Framework integration packages live under [`integrations/`](integrations/), orga
 framework-first then language: `integrations/<framework>/<language>/`. Each package is built,
 tested, and released independently.
 
-- **Python**: Google ADK, Microsoft Agent Framework, Microsoft AutoGen, AG2, CrewAI, LangGraph, LiveKit, Pydantic AI
+- **Python**: Google ADK, Microsoft Agent Framework, Microsoft AutoGen, AG2, CrewAI, LangGraph, LiveKit, Pydantic AI, Strands Agents
 - **TypeScript**: Google ADK, Mastra, Vercel AI SDK
 - **Go**: Google ADK
 

diff --git a/integrations/CLAUDE.md b/integrations/CLAUDE.md
@@ -30,6 +30,7 @@ Current packages:
 | `mastra/typescript` | `@getzep/zep-mastra` (npm) | `@getzep/zep-mastra` |
 | `ms-agent-framework/python` | `zep-ms-agent-framework` (PyPI) | `zep_ms_agent_framework` |
 | `pydantic-ai/python` | `zep-pydantic-ai` (PyPI) | `zep_pydantic_ai` |
+| `strands/python` | `zep-strands` (PyPI) | `zep_strands` |
 | `vercel-ai/typescript` | `@getzep/zep-vercel-ai` (npm) | `@getzep/zep-vercel-ai` |
 
 **Naming convention (keep CI derivation simple):** the framework directory is the package

diff --git a/integrations/README.md b/integrations/README.md
@@ -23,6 +23,7 @@ New to Zep? Sign up at [getzep.com](https://www.getzep.com) and create an API ke
 | Microsoft Agent Framework | Python | `zep-ms-agent-framework` | [`ms-agent-framework/python/`](ms-agent-framework/python/) |
 | Microsoft AutoGen | Python | [`zep-autogen`](https://pypi.org/p/zep-autogen) | [`autogen/python/`](autogen/python/) |
 | Pydantic AI | Python | `zep-pydantic-ai` | [`pydantic-ai/python/`](pydantic-ai/python/) |
+| Strands Agents | Python | `zep-strands` | [`strands/python/`](strands/python/) |
 | Vercel AI SDK | TypeScript | `@getzep/zep-vercel-ai` | [`vercel-ai/typescript/`](vercel-ai/typescript/) |
 
 Looking for bulk data ingestion (Slack exports, documents, email, CSV/JSON records,

diff --git a/integrations/strands/python/CHANGELOG.md b/integrations/strands/python/CHANGELOG.md
@@ -0,0 +1,58 @@
+# Changelog
+
+## Unreleased
+
+### Added
+
+- Live agent integration test (`test_integration_full_lifecycle`) exercising
+  `Agent` + `MemoryManager` + `ZepMemoryStore` against Zep Cloud and OpenAI,
+  including cross-thread recall and `on_user_created` (gated on
+  `ZEP_API_KEY` + `OPENAI_API_KEY`). Store-only round-trip remains available
+  with just `ZEP_API_KEY`.
+
+### Fixed
+
+- `ZepMemoryStore.initialize()` no longer calls Zep. `Agent.__init__` is
+  synchronous, so Strands runs that hook on a throwaway event loop in a worker
+  thread; provisioning there drove the caller's `AsyncZep` client from a second
+  event loop and raised `RuntimeError: ... is bound to a different event loop`
+  whenever the caller had already awaited that client (the pattern the README
+  and example recommend: `ensure_user`, then hand the client to the store).
+  Provisioning now happens on the store's first search or write, which always
+  runs on the agent's own loop.
+- `ZepMemoryStore` now rejects `extraction=True` (or an `ExtractionConfig`) at construction unless the store is writable user-graph mode with both `user_id` and `thread_id`, so `MemoryManager` never schedules extraction that would raise on every cycle.
+- `add()` no longer truncates oversized `json` payloads. Slicing JSON strips its closing syntax, so the size guard produced a document Zep would reject; oversized `json` now raises a `ValueError` pointing at chunking. `text`/`message` payloads are still truncated with a warning.
+- Corrected the `provisioning` module docstring (it incorrectly referred to
+  `ZepContextProvider`).
+- `zep_search` no longer returns raw exception text to the model. SDK errors
+  can include URLs, identifiers, or response bodies; the tool now returns a
+  fixed `"Graph search failed."` string and logs only the exception type
+  (and `status_code` when present), with the full traceback at DEBUG.
+- The example and live agent test now pass an explicit `OpenAIModel`. Both
+  require `OPENAI_API_KEY` and document `strands-agents[openai]`, but neither
+  passed `model=` to `Agent`, so Strands silently fell back to its
+  `BedrockModel()` default and the run failed on missing AWS credentials. The
+  model ID is overridable via `OPENAI_MODEL` (default `gpt-5-mini`).
+
+### Changed
+
+- Documented that Strands' default extraction cadence is every **5 turns**, so conversation batches only reach Zep when the trigger fires (or on `flush()`), delaying graph building relative to turn-by-turn persistence — in addition to Zep's asynchronous ingestion after messages arrive.
+- Documented the failure-handling contract: Zep SDK errors propagate out of `search`/`add`/`add_messages` by design, because `MemoryManager` and `ExtractionCoordinator` own failure isolation (skip-and-log, `AggregateMemoryError`, and high-water-mark rollback for retry). Added tests pinning that behavior.
+- Example and live tests flush at the session boundary after `invoke_async`,
+  which `MemoryManager` requires to persist buffered turns on that path.
+
+## 0.1.0 (2026-08-01)
+
+### Added
+
+- `ZepMemoryStore` — a Strands Agents `MemoryStore` backed by Zep's temporal Context Graph.
+- `search` via `graph.search` (default `scope="auto"` returns Zep's assembled Context Block).
+- `add_messages` via `thread.add_messages` for server-side extraction in user-graph mode.
+- `add` via `graph.add` for text/JSON/message facts (`metadata["type"]` selects the data type).
+- `initialize` provisions the Zep user and thread (no-op for standalone graphs).
+- Standalone-graph mode via `graph_id` (search + add only).
+- `ensure_user` / `ensure_thread` idempotent provisioning helpers with optional `on_created` hook.
+- `create_zep_search_tool` with pin-or-expose control over `graph.search` parameters.
+- `expose_search_tool` on `ZepMemoryStore` to register the search tool through `get_tools()`.
+- Message and graph payload truncation helpers with length-only warnings.
+- Mock-based unit tests, a gated live integration test, and a runnable example.
diff --git a/integrations/strands/python/Makefile b/integrations/strands/python/Makefile
@@ -0,0 +1,72 @@
+# Makefile for zep-strands development
+
+.PHONY: help install format lint lint-fix type-check test test-cov clean build all pre-commit ci
+
+# Default target
+help:
+	@echo "Available commands:"
+	@echo "  install     - Install package and dependencies in development mode"
+	@echo "  format      - Format code with ruff"
+	@echo "  lint        - Run linting checks"
+	@echo "  type-check  - Run type checking with mypy"
+	@echo "  test        - Run tests"
+	@echo "  test-cov    - Run tests with coverage report"
+	@echo "  all         - Run format, lint, type-check, and test"
+	@echo "  build       - Build the package"
+	@echo "  clean       - Clean build artifacts"
+
+# Install package in development mode
+install:
+	uv sync --extra dev
+
+# Format code
+format:
+	uv run ruff format .
+
+# Run linting checks
+lint:
+	uv run ruff check .
+
+# Fix linting issues automatically
+lint-fix:
+	uv run ruff check --fix .
+
+# Run type checking
+type-check:
+	uv run mypy src/
+
+# Run tests (mock-based; integration tests skip without API keys)
+test:
+	uv run pytest tests/ -v
+
+# Run tests with coverage
+test-cov:
+	uv run pytest tests/ -v --cov=zep_strands --cov-report=term-missing --cov-report=xml
+
+# Run all checks (format first, then lint, then type-check, then test)
+all: format lint type-check test
+
+# Build the package
+build:
+	uv build
+
+# Clean build artifacts
+clean:
+	rm -rf dist/
+	rm -rf build/
+	rm -rf *.egg-info/
+	rm -rf .pytest_cache/
+	rm -rf .mypy_cache/
+	rm -rf .ruff_cache/
+	find . -type d -name __pycache__ -exec rm -rf {} +
+	find . -type f -name "*.pyc" -delete
+	rm -f coverage.xml
+	rm -f .coverage
+
+# Development workflow - run this before committing
+pre-commit: lint-fix format lint type-check test
+	@echo "All checks passed! Ready to commit."
+
+# CI workflow - strict checks without auto-fixing
+ci: lint type-check test
+	@echo "CI checks passed!"