Peter Steinberger
d4e1a790ab
refactor: share media dimension parsing
2026-04-19 05:11:58 +01:00
Peter Steinberger
34abb441f6
refactor: share provider plugin id filtering
2026-04-19 05:09:57 +01:00
Peter Steinberger
984ecd98ca
refactor: share legacy config migration pipeline
2026-04-19 05:06:58 +01:00
Peter Steinberger
528f296cfc
refactor: share acp identity construction
2026-04-19 05:04:19 +01:00
Peter Steinberger
45381135df
refactor: share conversation target normalization
2026-04-19 05:02:03 +01:00
Peter Steinberger
2a14f76964
refactor: share plugin validation diagnostics
2026-04-19 04:59:40 +01:00
Peter Steinberger
ca3e5ffd89
refactor: reduce subagent requester wrapper duplication
2026-04-19 04:52:51 +01:00
Peter Steinberger
9686e518bc
test: share media generation reset helpers
2026-04-19 04:48:52 +01:00
Peter Steinberger
5ca33f7cb4
refactor: share model resolve fallback lookup
2026-04-19 04:46:11 +01:00
Peter Steinberger
dfe2e81829
refactor: share provider replay hook params
2026-04-19 04:40:22 +01:00
Peter Steinberger
3eed321081
refactor: share model allowlist parsing
2026-04-19 04:34:51 +01:00
Peter Steinberger
2a35ea4f07
test: share pi embedded helper setup
2026-04-19 04:31:01 +01:00
Peter Steinberger
efda761724
refactor: share cron flat recovery
2026-04-19 04:28:03 +01:00
Peter Steinberger
c63d6bf508
refactor: reuse codex search config types
2026-04-19 04:25:12 +01:00
Peter Steinberger
bcbb3de760
test: reuse run attempt fixture
2026-04-19 04:22:05 +01:00
Peter Steinberger
590474a9a4
test: share compact session fixture
2026-04-19 04:19:35 +01:00
Peter Steinberger
10e14bd5be
test: reuse sanitize assistant fixture
2026-04-19 04:16:58 +01:00
Peter Steinberger
bfea6bebc9
test: share subagent cleanup lookup
2026-04-19 04:14:55 +01:00
Peter Steinberger
ab4eb5aa94
test: share anthropic cache payload fixture
2026-04-19 04:12:37 +01:00
Peter Steinberger
f5c49758fc
test: share gateway exec allowlist fixture
2026-04-19 04:10:19 +01:00
Peter Steinberger
394c7a2357
test: share exec approval disabled fixture
2026-04-19 04:07:54 +01:00
Peter Steinberger
91ad6c2739
test: share mcp cache tool turn helper
2026-04-19 04:05:28 +01:00
Peter Steinberger
04697eca88
refactor: share channel action params
2026-04-19 04:03:16 +01:00
Peter Steinberger
1908967cfa
test: share auth profile env cleanup
2026-04-19 04:00:36 +01:00
Peter Steinberger
f54cf74ef6
test: share BTW sanitized user assertion
2026-04-19 03:58:08 +01:00
Peter Steinberger
44166f7cfe
test: share live model switch params
2026-04-19 03:55:35 +01:00
Peter Steinberger
6a87d6e814
test: share model fallback probe assertions
2026-04-19 03:52:57 +01:00
Peter Steinberger
0f871664c5
test: share bootstrap heartbeat fixture
2026-04-19 03:49:35 +01:00
Peter Steinberger
0a5515297e
test: share skill auth config fixtures
2026-04-19 03:47:24 +01:00
Peter Steinberger
97a3089cec
test: share unsafe skill scan fixture
2026-04-19 03:44:29 +01:00
Peter Steinberger
555f74cf67
test: share escaped bundled skill fixture
2026-04-19 03:42:19 +01:00
Peter Steinberger
9e93aa0c32
test: share ClawHub skill update assertion
2026-04-19 03:40:13 +01:00
Peter Steinberger
bf5b6cba70
test: share usage accumulator fixtures
2026-04-19 03:37:36 +01:00
stain lu
24b915ed41
fix: surface preserved stale session totals ( #67695 ) (thanks @stainlu)
...
* fix(agents): preserve session totalTokens when provider omits usage data
Fixes #67667
When a provider (e.g. MiniMax via Anthropic endpoint) does not return
usage data in its API response, hasNonzeroUsage() is false and the
entire totalTokens update block in persistSessionAfterRun is skipped.
This resets totalTokens to undefined, causing /status to show 0%
context usage even after compaction has calculated real token counts.
The fix preserves the previous totalTokens value when the current run
has no usage data, marking it as stale (totalTokensFresh: false) so
display layers know it is from a prior run. This is strictly better
than null — the user sees the last known context usage instead of 0%.
* ci: retrigger after flaky gateway shutdown test
* test(agents): port totalTokens regression test to withTempSessionStore helper post-rebase
* fix(status): surface preserved stale session totals
* fix: surface preserved stale session totals (#67695 ) (thanks @stainlu)
---------
Co-authored-by: Ayaan Zaidi <hi@obviy.us >
2026-04-19 08:06:36 +05:30
Peter Steinberger
8233ca6401
test: share sandbox docker create fixture
2026-04-19 03:35:20 +01:00
Peter Steinberger
bf2fbf071b
test: share vertex ADC auth fixture
2026-04-19 03:32:49 +01:00
Peter Steinberger
199f4d78d9
test: share anthropic payload fixtures
2026-04-19 03:29:43 +01:00
stain lu
4da808da50
fix: scope nested agent lanes per target session ( #67785 ) (thanks @stainlu)
...
* fix(agents): scope nested lane per target session to stop cross-agent blocking
* docs(agents): note per-session nested-lane lifecycle parity with session:* lanes
* refactor(agents): distill nested lane helpers
* fix: scope nested agent lanes per target session (#67785 ) (thanks @stainlu)
---------
Co-authored-by: Ayaan Zaidi <hi@obviy.us >
2026-04-19 07:58:55 +05:30
Peter Steinberger
67bd9edd8b
test: share cache trace memory fixture
2026-04-19 03:27:05 +01:00
Peter Steinberger
6de5f92835
test: share command delivery media fixture
2026-04-19 03:24:43 +01:00
Peter Steinberger
83a0f1fd52
test: share subagent cleanup decision fixture
2026-04-19 03:22:10 +01:00
Peter Steinberger
314654bd0f
test: share auth profile env fixture
2026-04-19 03:19:57 +01:00
Peter Steinberger
22d99ee9df
test: share models config env fixture
2026-04-19 03:17:36 +01:00
Peter Steinberger
8f92c0607c
test: share transcript replay defaults fixture
2026-04-19 03:15:23 +01:00
Ayaan Zaidi
74f0dc87de
fix: always send openai stream usage flag ( #68746 ) (thanks @kagura-agent)
2026-04-19 07:44:48 +05:30
Ayaan Zaidi
43f6ffd0ae
test: distill openai stream usage regression coverage
2026-04-19 07:44:48 +05:30
kagura-agent
c560793482
fix: always send stream_options.include_usage when streaming openai-completions
...
Backends like llama-cpp and LM Studio require stream_options: { include_usage: true }
in the request payload to report token usage in streaming responses.
buildOpenAICompletionsParams() previously gated this behind supportsUsageInStreaming
compat detection, which excluded non-standard and custom endpoints. The OpenAI SDK
sends this unconditionally, so we now do the same.
Fixes #68707
2026-04-19 07:44:48 +05:30
Peter Steinberger
1212412ff1
test: share context window model fixture
2026-04-19 03:12:59 +01:00
Peter Steinberger
a56aa6ccbe
test: share model compat streaming fixture
2026-04-19 03:10:47 +01:00
Peter Steinberger
59032f63b1
test: share compact skill prompt fixture
2026-04-19 03:08:24 +01:00
