 opieter-aws
|
79bfd80a85
|
docs: add Observability catalog type (#4772)
|
2026-10-01 10:04:16 -04:00 |
|
|
opieter-aws
|
ea7305f51e
|
docs(agents): add note on invocation state (#4744)
|
2026-09-30 15:31:59 -04:00 |
|
|
opieter-aws
|
c73253db35
|
fix(harness): remove context offloader plugin from harness in favor of context manager's stash (#4701)
|
2026-09-30 10:51:56 -04:00 |
|
|
opieter-aws
|
db34d016a8
|
fix(context-manager): keep stashed originals when backfilling after session resume (#4699)
|
2026-09-29 17:00:36 -04:00 |
|
|
opieter-aws
|
b7900ae688
|
docs(harness): update default model (#4661)
|
2026-09-28 15:03:04 -04:00 |
|
|
opieter-aws
|
b660f0e594
|
fix(caching): surface prompt-cache write tokens on TS OpenAI and Bedrock paths (#4617)
|
2026-09-28 09:21:42 -04:00 |
|
|
opieter-aws
|
219dbf6587
|
ci: add Python lint/format gate to pre-commit hook (#4528)
|
2026-09-23 12:27:10 -04:00 |
|
|
opieter-aws
|
fec0427664
|
feat(agent): add Agent.shutdown() for scope-based cleanup (#4519)
|
2026-09-22 19:14:05 -04:00 |
|
|
opieter-aws
|
7462908f43
|
feat: automatically use openAI prompt-cache keys from session id (#4083)
|
2026-09-10 10:54:22 -04:00 |
|
|
opieter-aws
|
5f636bbc2b
|
fix(openai): emit tool results before user text to preserve tool_use/tool_result adjacency (#4234)
|
2026-09-08 14:49:52 -04:00 |
|
|
opieter-aws
|
42aedfc034
|
feat: accept cache_config on Ollama, Writer, and SageMaker (#4232)
|
2026-09-08 12:14:01 -04:00 |
|
|
opieter-awsandStrandly
|
3b4ca85d21
|
fix(vercel): emit tool results before user text to preserve tool_use/tool_result adjacency (#4138)
Co-authored-by: Strandly (The Agent) <strands-agents+strandly@amazon.com>
|
2026-09-08 08:25:45 -04:00 |
|
|
opieter-aws
|
55803bc7e5
|
fix: gate the deprecated cache_tools point on the resolved cache strategy (#4186)
|
2026-09-04 16:06:51 -04:00 |
|
|
opieter-aws
|
24bc5c3d5b
|
feat!: require Node.js 22+, drop Node 20 support (#4145)
|
2026-09-03 16:03:51 -04:00 |
|
|
opieter-aws
|
9be952fde7
|
feat: add cache config to llama.cpp (#4127)
|
2026-09-03 08:36:49 -04:00 |
|
|
opieter-aws
|
022e5a51a9
|
ci: cancel superseded PR runs to reduce runner concurrency (#4134)
|
2026-09-03 08:18:48 -04:00 |
|
|
opieter-aws
|
a02c1db8c0
|
feat: support caching for vercel provider (#4055)
|
2026-09-03 08:03:55 -04:00 |
|
|
opieter-aws
|
e13bfcd4c7
|
test(bedrock): stabilize the cache-point-after-document integ test (#4119)
|
2026-09-02 16:48:09 -04:00 |
|
|
opieter-aws
|
33f139b704
|
fix: include cached tokens in input_tokens (#4080)
|
2026-09-01 08:22:21 -04:00 |
|
|
opieter-aws
|
eabcb137ad
|
feat(models): add cache_config to Mistral (#3986)
|
2026-08-28 11:20:02 -04:00 |
|
|
opieter-aws
|
a6c2e5e812
|
fix: count cached tokens in context-size and compaction baseline (#3886)
|
2026-08-27 14:13:48 -04:00 |
|
|
opieter-aws
|
5cadef436a
|
fix(telemetry): emit semconv-compliant cache usage attributes (#3964)
|
2026-08-25 14:03:26 -04:00 |
|
|
opieter-aws
|
f34de8c088
|
feat(models): add cache_key to CacheConfig for key-routed cache providers (#3949)
|
2026-08-25 10:45:26 -04:00 |
|
|
opieter-aws
|
91d527496a
|
fix: harden mantle routing integ tests (#3957)
|
2026-08-25 14:40:32 +00:00 |
|
|
opieter-aws
|
7b42cf9df2
|
fix: harden vercel document test (#3958)
|
2026-08-24 17:16:12 -04:00 |
|
|
opieter-aws
|
25a8ae6cca
|
feat(memory-ts): export key-aware extractor and make it model-configurable (#3948)
|
2026-08-24 15:47:54 -04:00 |
|
|
opieter-aws
|
64d54a1281
|
test: harden bedrock kb integ tests (#3954)
|
2026-08-24 15:46:35 -04:00 |
|
|
opieter-aws
|
36289b8dba
|
fix(models): warn and skip cachePoint message blocks instead of raising (#3947)
|
2026-08-24 11:45:26 -04:00 |
|
|
opieter-aws
|
aac15db547
|
fix: count Gemini tool-use tokens as input and thinking tokens as output (#3892)
|
2026-08-21 12:53:11 -04:00 |
|
|
opieter-aws
|
606e1b84dc
|
fix(multiagent): accumulate cache token counters in Graph and Swarm (#3884)
|
2026-08-20 07:28:26 -04:00 |
|
|
opieter-aws
|
f79ce81011
|
test(openai): remove unwinnable rate-limit throttling integ test (#3891)
|
2026-08-20 07:22:56 -04:00 |
|
|
opieter-aws
|
2b6af5a1da
|
fix: fix failing unit tests for bedrock caching (#3887)
|
2026-08-19 15:46:51 -04:00 |
|
|
opieter-aws
|
2b2d9bd40f
|
fix: read prompt_tokens_details cache reads in TS OpenAI Chat (#3885)
|
2026-08-19 11:25:11 -04:00 |
|
|
opieter-aws
|
8bd807dfc9
|
fix(bedrock): inherit cache_config ttl on an untimed tools cache point (#3858)
|
2026-08-17 15:09:54 -04:00 |
|
|
opieter-aws
|
cc99e8c15e
|
docs: add note for disabling conversation managers (#3823)
|
2026-08-14 14:52:45 -04:00 |
|
|
opieter-aws
|
d2567ce5ec
|
feat: add injected content behind cache points (#3704)
|
2026-08-14 11:35:38 -04:00 |
|
|
opieter-aws
|
5b3e060777
|
test: add integration testing for caching (#3793)
|
2026-08-13 15:55:28 -04:00 |
|
|
opieter-aws
|
1c45a8b504
|
ci(anthropic): re-enable anthropic integration tests (#3783)
|
2026-08-12 13:35:26 -04:00 |
|
|
opieter-aws
|
7741ecd286
|
fix(swarm): correct crash-restart resume for handoff and completed swarms (#3391)
|
2026-08-06 13:21:32 -04:00 |
|
|
opieter-aws
|
a059fb91ba
|
feat: add snapshot session manager to python (#3283)
|
2026-08-05 15:37:34 -04:00 |
|
|
opieter-awsandstrandly-the-agent
|
0ab31d7c12
|
feat: support Tool Choice for Gemini in Python (#3551)
Co-authored-by: strandly-the-agent <strandly-the-agent@users.noreply.github.com>
|
2026-08-04 11:40:24 -04:00 |
|
|
opieter-aws
|
6f6fb65d19
|
docs(memory): add TestMemoryStore documentation (#3199)
|
2026-07-23 17:13:25 -04:00 |
|
|
opieter-aws
|
ba37a272c9
|
fix(multiagent): preserve shared context and cumulative accounting across serialize/deserialize (#3396)
|
2026-07-23 14:25:14 -04:00 |
|
|
opieter-aws
|
81bccd8420
|
fix: verify aws region (#3412)
|
2026-07-22 16:18:18 -04:00 |
|
|
opieter-aws
|
33f5f74116
|
fix: reject keys for s3 storage if not configured (#3411)
|
2026-07-22 14:44:54 -04:00 |
|
|
opieter-aws
|
b0ff5ef3f5
|
fix(graph): keep fan-in node out of resume while a parallel sibling is in-flight (#3390)
|
2026-07-22 09:58:53 -04:00 |
|
|
opieter-aws
|
41ccf5e106
|
refactor(memory): refactor TestMemoryStore to use the unified storage interface (#3260)
|
2026-07-21 12:14:08 -04:00 |
|
|
opieter-aws
|
ad5685f7e4
|
docs(memory): add community page for AgentCore Memory Store (#3316)
|
2026-07-20 11:04:06 -04:00 |
|
|
opieter-aws
|
c613969c42
|
docs: tweak pr-writer skill to be more concise (#3148)
|
2026-07-09 11:22:28 -04:00 |
|
|
opieter-aws
|
6c58f97ca6
|
fix(memory)!: rename LocalMemoryStore to TestMemoryStore (#3123)
|
2026-07-09 11:04:16 -04:00 |
|
|
opieter-aws
|
818a985942
|
refactor(event_loop): route message appends through Agent._append_messages (#3131)
|
2026-07-09 10:49:59 -04:00 |
|
|
opieter-aws
|
6993ea3f84
|
docs: improve agent guidance on issue references in regression tests (#3146)
|
2026-07-09 09:32:12 -04:00 |
|
|
opieter-aws
|
c4d27dc698
|
docs(memory): add memory stores section on community page (#3130)
|
2026-07-09 09:31:42 -04:00 |
|
|
opieter-aws
|
a1915b6c62
|
feat: add durable identifiers to messages (#2836)
|
2026-07-08 15:18:11 -04:00 |
|
|
opieter-aws
|
a2c3c050fd
|
ci: fix integration test runs (#3112)
|
2026-07-07 16:14:30 +00:00 |
|
|
opieter-aws
|
2349304de3
|
docs: improve agent guidance (#2959)
|
2026-06-26 11:34:39 -04:00 |
|
|
opieter-aws
|
84fff9a908
|
feat: add local memory store (#2859)
|
2026-06-26 11:00:29 -04:00 |
|
|
opieter-aws
|
d9b9061486
|
feat: telemetry for memory manager (#2858)
|
2026-06-25 17:05:11 -04:00 |
|
|
opieter-aws
|
bbaee8b441
|
ci: skip SDK test matrices on markdown-only changes (#2967)
|
2026-06-25 19:33:54 +00:00 |
|
|
opieter-aws
|
188d474981
|
fix: run bedrock-kb store test in node only (#2966)
|
2026-06-25 15:01:04 -04:00 |
|
|
opieter-aws
|
90cbef6c4f
|
docs: improve testing instructions (#2965)
|
2026-06-25 14:00:46 -04:00 |
|
|
opieter-aws
|
83e70aceb5
|
docs: add TESTING.md for python (#2961)
|
2026-06-25 11:51:11 -04:00 |
|
|
opieter-aws
|
47c76fb379
|
docs: add memory manager docs (#2758)
|
2026-06-17 13:00:59 +00:00 |
|
|
opieter-aws
|
febaed9947
|
fix: mark internal memory functions as private (#2817)
|
2026-06-16 08:54:33 -04:00 |
|
|
opieter-aws
|
6f9cae056a
|
feat: add default for memory extraction trigger (#2811)
|
2026-06-15 22:16:30 -04:00 |
|
|
opieter-aws
|
a111c5d2dd
|
feat: add memory injection (#2797)
|
2026-06-15 17:37:41 -04:00 |
|
|
opieter-aws
|
675e35c372
|
chore: add CLAUDE.md files (#2809)
|
2026-06-15 21:01:37 +00:00 |
|
|
opieter-aws
|
96fec40660
|
test: add integration test for memory manager (#2764)
|
2026-06-15 11:11:21 -04:00 |
|
|
opieter-aws
|
758675951f
|
ci: bump esbuild version (#2785)
|
2026-06-15 09:17:12 -04:00 |
|
|
opieter-aws
|
dacb784d43
|
feat: memory injection (#2631)
|
2026-06-12 13:02:11 -04:00 |
|
|
opieter-aws
|
13839973d3
|
feat: add extraction defaults and support for bedrock kbs (#2719)
|
2026-06-11 12:40:44 -04:00 |
|
|
opieter-aws
|
3a3fe7bc23
|
feat: pass sequenceNumber to memory store addMessages method (#2721)
|
2026-06-11 12:13:58 -04:00 |
|
|
opieter-aws
|
e21098a411
|
fix: improve devx for instantiating multiple bedrock KBs (#2722)
|
2026-06-11 08:04:09 -04:00 |
|
|
opieter-aws
|
7a2d4a7064
|
feat: add memory extraction (#2671)
|
2026-06-10 11:37:30 -04:00 |
|
|
opieter-awsandStrands Agent
|
e0c54fc9a4
|
docs: remove unnecessary model IDs from examples, rely on built-in defaults (#2645)
Co-authored-by: Strands Agent <217235299+strands-agent@users.noreply.github.com>
|
2026-06-05 16:05:00 -04:00 |
|
|
opieter-aws
|
8f4a8ebedc
|
fix: include json blocks in counting tokens (#2639)
|
2026-06-05 13:23:32 -04:00 |
|
|
opieter-aws
|
83f2ce7f79
|
chore: change memory store return type on add (#2613)
|
2026-06-03 12:18:45 -04:00 |
|
|
opieter-aws
|
1d9ce24dee
|
feat(memory): add memory manager (#2544)
|
2026-06-03 09:53:19 -04:00 |
|
|
opieter-aws
|
ccbeede1a6
|
design: add long term memory design (#844)
|
2026-06-01 17:00:39 -04:00 |
|
|
opieter-aws
|
b63c8dd349
|
ci: allow design type in PR title validation (#2395)
|
2026-06-01 10:15:43 -04:00 |
|
|
opieter-aws
|
c7755ee29a
|
docs: add proactive compression docs (#806)
|
2026-05-13 18:43:49 +00:00 |
|
|
|
1e4e4f2698
|
fix: default useNativeTokenCount to false (#1056)
Co-authored-by: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Co-authored-by: Strands Agent <217235299+strands-agent@users.noreply.github.com>
|
2026-05-13 13:44:56 +00:00 |
|
|
opieter-aws
|
305a0051b2
|
fix: set use_native_token_count default to false (#2284)
|
2026-05-13 09:09:03 -04:00 |
|
|
opieter-aws
|
1847faec4f
|
feat: cache AccessDenied error for count tokens (#2279)
|
2026-05-11 11:09:27 -04:00 |
|
|
opieter-aws
|
d99cd49707
|
feat: cache AccessDenied error for count tokens (#1032)
|
2026-05-08 20:04:37 +00:00 |
|
|
opieter-awsandagent-of-mkmeral
|
f8621853d3
|
feat: add proactive context compression to conversation managers (#2239)
Co-authored-by: agent-of-mkmeral <agent-of-mkmeral@users.noreply.github.com>
|
2026-05-08 10:02:44 -04:00 |
|
|
|
be309fe716
|
feat: add proactive context compression to conversation managers (#965)
Co-authored-by: agent-of-mkmeral <agent-of-mkmeral@users.noreply.github.com>
Co-authored-by: agent-of-mkmeral <217235299+strands-agent@users.noreply.github.com>
|
2026-05-07 17:54:23 +00:00 |
|
|
opieter-awsandLiz
|
5452be518d
|
design: add Context Offload plugin (#762)
Co-authored-by: Liz <91279165+lizradway@users.noreply.github.com>
|
2026-05-07 10:38:17 -04:00 |
|
|
opieter-aws
|
800e7c4661
|
feat: add useNativeTokenCount flag to skip token counting API calls (#2255)
|
2026-05-06 16:51:04 -04:00 |
|
|
opieter-aws
|
ed32a3007c
|
feat: add useNativeTokenCount flag to skip token counting API calls (#1009)
|
2026-05-06 20:50:47 +00:00 |
|
|
opieter-aws
|
b21144c032
|
fix: cache unsupported models for bedrocks token counting (#999)
|
2026-05-06 16:11:46 +00:00 |
|
|
opieter-aws
|
6b0df9add0
|
fix: cache unsupported models for bedrocks token counting (#2250)
|
2026-05-06 11:41:20 -04:00 |
|
|
opieter-aws
|
559b2a0572
|
feat: add context window limit lookup table (#2249)
|
2026-05-05 11:44:22 -04:00 |
|
|
opieter-aws
|
a12ea3e3c4
|
fix: run generate before type-check in strands-wasm (#987)
|
2026-05-01 20:46:06 +00:00 |
|
|
opieter-aws
|
7c00a77aa6
|
feat: auto-populate contextWindowLimit from model ID lookup tables (#954)
|
2026-05-01 15:23:00 +00:00 |
|
|
opieter-aws
|
888c98c430
|
feat: estimate input tokens before model calls (#2221)
|
2026-04-29 14:19:14 -04:00 |
|
|
opieter-aws
|
52cdb9da0b
|
fix: change token counting fallback log from warning to debug (#2220)
|
2026-04-28 14:22:22 -04:00 |
|
|
opieter-aws
|
d5611a8c31
|
fix: change token counting fallback log from warn to debug (#942)
|
2026-04-28 18:16:05 +00:00 |
|
|
opieter-aws
|
03f3c53dae
|
fix: remove internal ProviderTokenCountError from public exports (#937)
|
2026-04-28 15:04:11 +00:00 |
|
|
opieter-aws
|
009374f483
|
feat: add ProviderTokenCountError for native token counting failures (#2211)
|
2026-04-28 10:54:11 -04:00 |
|