devin-ai-integration[bot] 9d16412341 fix(caching): count tool_call cache_control marks in the injection census (#43556) ... * fix(caching): count tool_call cache_control marks in the injection census Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(caching): remove cache census casts Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(caching): only skip injection on message or content marks Co-Authored-By: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> * fix(caching): skip injection on messages whose tool calls carry marks Reverts 9f08d8aef8. A default 5m mark injected on assistant text lands before the client's 1h tool_use mark, which Anthropic rejects with a 400 because a 1h breakpoint must not follow a 5m one. Keeping the full census in the skip check leaves the client's tool_call breakpoint as the only one on that message. * fix(caching): count every client tool_call cache mark in the breakpoint census The census gated tool_call marks on type function and dict shape, so a client mark on a call without a type or with a string cache_control slipped past the count and injection overflowed the 4 breakpoint cap. Count any non-None tool_call mark, keep the server tool exclusion, and add integration cells for the capped surfaces, the yaml stand-down, Bedrock and Gemini, and router affinity * test(integration): hold the upstream so the worker kill lands mid-burst * refactor(caching): reuse the transform's server tool lookup in the breakpoint census The census now calls the same helper the Anthropic transform uses to decide whether a marked tool call becomes a server tool block, so the two cannot drift apart. The owned-proxy burst test waits up to 90 seconds for the burst to reach the wire before it kills a worker * refactor(anthropic): move the server tool rebuild check under llms/anthropic * test(integration): audit the tool call mark census across chat, messages, responses, and chaos Adds the /audit cells for the breakpoint census on assistant tool_calls marks: the Responses stream bridge, the OpenAI and Anthropic SDK clients, in-process Pydantic messages, response cache twins, request-level points, a provider 401 on a capped request, malformed provider_specific_fields and tool_call ids, null or empty points, a points update mid-burst, a proxy restart mid-burst, and a worker SIGKILL that picks the worker holding the burst's upstream connections * test(integration): close the SDK clients the cache census cells open --------- Co-authored-by: Devin AI <158243242+devin-ai-integration[bot]@users.noreply.github.com> Co-authored-by: mateo-berri <277851410+mateo-berri@users.noreply.github.com>		2026-10-03 17:33:48 -07:00
..
a2a
access_control
batches	test(proxy): delete the legacy proxy test tree and shard tests/unit/proxy by glob (#44018)	2026-10-01 11:40:45 -07:00
claude_code	feat(e2e): read management routes back from the control plane replicas (#43373)	2026-09-26 18:05:55 -07:00
coverage_registry	fix(guardrails): run unified guardrails on /v1/images/edits (#44195)	2026-10-02 21:14:35 -07:00
gateway	test(e2e): add secret manager lanes for HashiCorp Vault and CyberArk Conjur (#42503)	2026-09-22 18:02:32 -07:00
guardrails	fix(guardrails): run unified guardrails on /v1/images/edits (#44195)	2026-10-02 21:14:35 -07:00
llm_translation	fix(caching): count tool_call cache_control marks in the injection census (#43556)	2026-10-03 17:33:48 -07:00
load	feat(lint): add LIT013 flagging *-ok suppressions that suppress nothing and remove the 240 stale ones (#42793)	2026-09-23 17:50:09 -07:00
logging	fix(otel): name postgres service spans by operation and table (#44240)	2026-10-03 09:20:28 -07:00
management	fix(proxy): delete large teams without per-member transaction fan-out (#42998)	2026-09-30 13:49:18 -07:00
mcp	feat(jwt): auto_register_map_existing_key maps JWT to the user's existing virtual key (#42375)	2026-10-02 12:11:33 -07:00
migrations	feat(lens): coordinate worker releases and bundled installs (#44428)	2026-10-03 16:33:45 -07:00
other	feat(jwt): auto_register_map_existing_key maps JWT to the user's existing virtual key (#42375)	2026-10-02 12:11:33 -07:00
quota_management	test(proxy): move middleware, spend_tracking, pass_through, common_utils and root proxy tests into tests/unit/proxy (#44015)	2026-10-01 18:23:31 +00:00
router	test(e2e): keep 1ms-timeout deployments off the provider cache (#44082)	2026-10-01 15:54:02 -07:00
secret_manager	fix(ci): stop stale CI reds, keep unit tests off the host env, retry CyberArk policy conflicts (#43294)	2026-09-26 09:25:13 -07:00
ui	refactor(ui): compose dashboard pages with shared layouts (#44306)	2026-10-03 01:35:34 +00:00
AGENTS.md	test(e2e): typed per-test metadata for the e2e suite (#42044)	2026-09-30 21:03:21 -07:00
conftest.py	feat(jwt): auto_register_map_existing_key maps JWT to the user's existing virtual key (#42375)	2026-10-02 12:11:33 -07:00
CONTRIBUTING.md	fix(auth): give UI/CLI session tokens their own AES-GCM context and header-safe shape (#43790)	2026-09-29 18:42:24 -07:00
e2e_config.py	feat(jwt): auto_register_map_existing_key maps JWT to the user's existing virtual key (#42375)	2026-10-02 12:11:33 -07:00
e2e_db.py
e2e_http.py	feat(vertex): native batch JSONL passthrough with cost tracking (#42810)	2026-09-24 12:35:34 -07:00
e2e_metadata.py	test(e2e): typed per-test metadata for the e2e suite (#42044)	2026-09-30 21:03:21 -07:00
fixture_bundle.py
fixture_canonical.py
fixture_mode.py
fixture_profile.py
idp.py
idp_realm.json
junit_properties.py	test(e2e): typed per-test metadata for the e2e suite (#42044)	2026-09-30 21:03:21 -07:00
lifecycle.py	feat(lint): add LIT013 flagging *-ok suppressions that suppress nothing and remove the 240 stale ones (#42793)	2026-09-23 17:50:09 -07:00
memory_readings.py
models.py	fix(caching): count tool_call cache_control marks in the injection census (#43556)	2026-10-03 17:33:48 -07:00
otel_client.py	test(e2e): accept the otel cost write as a linked root trace (#42931)	2026-09-25 22:40:45 -07:00
PROVIDER_CACHE.md
provider_cache.py
provider_cache_redis.py
provider_cache_routing.py
provider_edge.py	fix(proxy): send a real error event when a /v1/messages stream fails (#41826)	2026-09-25 15:36:35 -07:00
provider_edge_bedrock.py
proxy_client.py	feat(e2e): record each e2e test's steps, starting with ProxyClient (#42393)	2026-09-30 19:33:53 -07:00
pytest.ini	feat(jwt): auto_register_map_existing_key maps JWT to the user's existing virtual key (#42375)	2026-10-02 12:11:33 -07:00
stack_lock.py	test(e2e): run the memory cell alone on the shared stack (#42518)	2026-09-22 15:13:37 -07:00
test_e2e_http.py	test(e2e): align completion, SAIL, and spend-log fixtures with supported contracts (#43902)	2026-09-30 14:21:22 -07:00
test_fixture_bundle.py
test_fixture_canonical.py
test_fixture_mode.py
test_idp.py
test_junit_properties.py
test_provider_edge.py	fix(proxy): send a real error event when a /v1/messages stream fails (#41826)	2026-09-25 15:36:35 -07:00
test_proxy_client.py	feat(e2e): read management routes back from the control plane replicas (#43373)	2026-09-26 18:05:55 -07:00
test_stack_lock.py	test(e2e): run the memory cell alone on the shared stack (#42518)	2026-09-22 15:13:37 -07:00
transport.py