strands-agents/harness-sdk python/v1.43.0 on GitHub

What's Changed

feat(strands-py): add optional hook order by @lizradway in #2559
feat(memory): add memory manager by @opieter-aws in #2544
chore: change memory store return type on add by @opieter-aws in #2613
feat(docs): add human in the loop by @lizradway in #2558
feat(checkpoint): wire checkpointing into agent event loop by @JackYPCOnline in #2190
feat(strands-py-wasm): make get/set app state sync by @pgrayy in #2610
feat(strands-py-wasm): wire vended tools through the WASM guest by @pgrayy in #2607
docs: remove archived agent-builder references by @zastrowm in #2616
fix: surface MaxTokensError when max_tokens truncates tool input JSON by @JackYPCOnline in #2620
fix: add maintain role to auto-strands-review allowed-roles by @zastrowm in #2633
feat: simplify DockerSandbox to bring-your-own-container by @gautamsirdeshmukh in #2561
chore(docs): remove duplicate community callouts from xAI provider page by @zastrowm in #2637
feat: add AgentCore tool search community plugin page by @senthilkumarmohan in #2623
chore: update repository references to harness-sdk by @zastrowm in #2618
fix: include json blocks in counting tokens by @opieter-aws in #2639
chore(site): update GitHub dropdown to harness-sdk and add evals by @zastrowm in #2641
docs: remove unnecessary model IDs from examples, rely on built-in defaults by @opieter-aws in #2645
feat: add test-infra/ CDK stack for long-running integ test resources by @notowen333 in #2650
docs: document per-invocation limits feature by @notowen333 in #2638
docs(strands-ts): update Bedrock examples to default model id by @JackYPCOnline in #2669
chore: add missing policies to integ test role by @notowen333 in #2670
chore: fixup test-infra constructs by @notowen333 in #2673
fix(strands-ts): update package.json repository URLs to harness-sdk by @zastrowm in #2675
feat(strands-py): add Sandbox core abstraction (TS→Python port, core only 1/N) by @agent-of-mkmeral in #2665
feat(context): add context_manager="auto" facade on Agent by @lizradway in #2643
refactor(site): use local monorepo source instead of cloning SDKs by @zastrowm in #2677
feat: add claude-opus-4-8 to context window limits by @lizradway in #2676
fix(a2a): introduce agent factory for isolating agent context from different callers by @JackYPCOnline in #2628
feat: add model_state as a snapshot field by @zastrowm in #2680
fix(site): update community contribution link to harness-sdk repo by @zastrowm in #2695
refactor(a2a): use native model_state snapshot field by @JackYPCOnline in #2694
fix: remove orphaned evaluation stub page by @zastrowm in #2678
feat(site): generate smart preview links for changed doc pages by @zastrowm in #2692
fix(site): update stale GitHub links from old repos to harness-sdk by @zastrowm in #2698
ci: add explicit permissions to workflow jobs by @yonib05 in #2388
fix: address CodeQL static analysis findings in JS/TS by @yonib05 in #2390
docs: add steering intervention handler documentation for TypeScript by @lizradway in #2624
ci: add automated issue labeling workflow by @yonib05 in #2359
feat(context): add message pinning to conversation managers by @lizradway in #2644
fix: update remaining sdk-python references to harness-sdk by @zastrowm in #2701
feat: add Bedrock Knowledge Base memory store by @notowen333 in #2630
ci: bump actions/checkout from 4 to 6 by @dependabot[bot] in #2100
ci: add PR labeling to issue-labeler workflow by @yonib05 in #2702
feat(python): add Docker/SSH Sandbox implementations by @gautamsirdeshmukh in #2691
fix(bedrock): preserve signature-only reasoning blocks across turns by @pgrayy in #2704
feat: add pr-create and pr-writer agent skills by @zastrowm in #2646
feat: add pr-feedback agent skill for triaging PR review comments by @zastrowm in #2712
feat: add memory extraction by @opieter-aws in #2671
docs: add README to .agents/skills/ explaining each skill's purpose by @zastrowm in #2714
blog: inside Agentic Football Cup — 10 Strands Agents, 1 Second to Decide by @ianholtz in #2651
feat: add middleware system for wrapping agent stages by @zastrowm in #2681
docs(a2a): document per-context conversation isolation by @JackYPCOnline in #2699
fix(a2a): isolate conversation state per context in TypeScript by @JackYPCOnline in #2696
feat: migrate TS integration test role to shared test-infra CDK stack by @notowen333 in #2715
ci: allow blog type in PR title validation by @zastrowm in #2718
fix(strands-py): default to us-east-1 for integration tests by @zastrowm in #2720
ci: enforce API review label requirement before merge by @zastrowm in #2716
feat: allow same-account IAM roles to assume the integ test role by @notowen333 in #2723
fix: improve devx for instantiating multiple bedrock KBs by @opieter-aws in #2722
fix(dependabot): label npm updates typescript instead of javascript by @yonib05 in #2726
fix(template): remove stale consuming-repos reference from PR template by @yonib05 in #2727
feat: integrate Sandbox with Agent by @gautamsirdeshmukh in #2563
feat: update vended tools/plugins for sandbox compatibility by @gautamsirdeshmukh in #2649
feat: pass sequenceNumber to memory store addMessages method by @opieter-aws in #2721
feat: add extraction defaults and support for bedrock kbs by @opieter-aws in #2719
feat: implement intervention primitive in python with cancellation support by @mehtarac in #2693
fix(strands-py): remove flaky test_graph_parallel_execution test by @lizradway in #2743
fix(issue-labeler): stop over-applying area-community and chore by @yonib05 in #2725
docs: add AI contribution guidance to CONTRIBUTING and PR template by @yonib05 in #2728
docs(agents): add community and character guidance to AGENTS.md by @yonib05 in #2747
docs(agents): add PR quality bar to AGENTS.md by @yonib05 in #2729
docs: add context management page for context_manager=auto by @lizradway in #2741

New Contributors

@senthilkumarmohan made their first contribution in #2623
@ianholtz made their first contribution in #2651

Full Changelog: python-wasm/v0.0.1...python/v1.43.0