fix(core): reject truncated subagent write_file calls - #3505
Merged
Conversation
Propagate MAX_TOKENS truncation from subagent responses into tool requests and reject truncated edit calls before schema validation can surface misleading missing-parameter errors. Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
Contributor
Code Coverage Summary
CLI Package - Full Text ReportCore Package - Full Text ReportFor detailed HTML reports, please see the 'coverage-reports-22.x-ubuntu-latest' artifact from the main CI run. |
When a streaming response hits MAX_TOKENS and is retried, accumulated state variables (functionCalls, wasOutputTruncated, roundText, etc.) were not cleared. This caused a successful retry to inherit the stale wasOutputTruncated=true flag from the failed attempt, incorrectly rejecting all Edit tool calls with a truncation error. Reset all per-attempt state on retry events, matching the existing behaviour in turn.ts. Add a regression test covering the truncated-then-retried-successfully scenario. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code)
doudouOUC
added
type/feature-request
doudouOUC
requested review from
LaZzyMan,
jifeng,
pomelo-nwu,
tanzhenxin and
wenshao
April 21, 2026 17:02
wenshao
approved these changes
Apr 21, 2026
wenshao
left a comment
wenshao
left a comment
Collaborator
There was a problem hiding this comment.
No issues found. LGTM! ✅ — gpt-5.4 via Qwen Code /review
Collaborator
|
Took this for a thorough verification run before merge — local test suite (5896 tests) is green, typecheck/build clean, both new tests pass. One small coverage suggestion before merge if you have a moment: The existing test Without it('should discard pre-retry function calls so only the retried call executes', async () => {
const writeFileToolDef: FunctionDeclaration = {
name: WriteFileTool.Name,
description: 'Writes a file',
parameters: { type: Type.OBJECT, properties: {} },
};
const { config } = await createMockConfig({
getFunctionDeclarationsFiltered: vi
.fn()
.mockReturnValue([writeFileToolDef]),
getTool: vi.fn().mockImplementation((name: string) => {
if (name === WriteFileTool.Name) {
return new WriteFileTool(config);
}
return undefined;
}),
});
const toolConfig: ToolConfig = { tools: [WriteFileTool.Name] };
const toolResultEvents: AgentToolResultEvent[] = [];
const eventEmitter = new AgentEventEmitter();
eventEmitter.on(AgentEventType.TOOL_RESULT, (event: unknown) => {
toolResultEvents.push(event as AgentToolResultEvent);
});
// Round 1: emit a (would-be) tool call, then a retry, then a different
// tool call. The retry handler must reset the accumulated functionCalls
// array — otherwise BOTH calls execute and the pre-retry one (which
// was supposed to be discarded) leaks into tool execution.
let callCount = 0;
mockSendMessageStream.mockImplementation(async () => {
callCount++;
if (callCount === 1) {
return (async function* () {
yield {
type: 'chunk',
value: {
functionCalls: [
{
id: 'call_pre_retry',
name: WriteFileTool.Name,
args: {
file_path: '/tmp/pre-retry.txt',
content: 'should-not-run',
},
},
],
},
};
yield { type: 'retry' };
yield {
type: 'chunk',
value: {
functionCalls: [
{
id: 'call_post_retry',
name: WriteFileTool.Name,
args: {
file_path: '/tmp/post-retry.txt',
content: 'should-run',
},
},
],
},
};
yield {
type: 'chunk',
value: {
candidates: [
{ finishReason: 'STOP', content: { parts: [] } },
],
},
};
})();
}
return (async function* () {
yield {
type: 'chunk',
value: {
candidates: [
{
finishReason: 'STOP',
content: { parts: [{ text: 'done' }] },
},
],
},
};
})();
});
const scope = await AgentHeadless.create(
'test-agent',
config,
promptConfig,
defaultModelConfig,
defaultRunConfig,
toolConfig,
eventEmitter,
);
await scope.execute(new ContextState());
const writeResults = toolResultEvents.filter(
(event) => event.name === WriteFileTool.Name,
);
// Exactly one write_file invocation should execute — the retried one.
expect(writeResults).toHaveLength(1);
expect(writeResults[0].callId).toBe('call_post_retry');
// The discarded call must never have produced a TOOL_RESULT event.
expect(
toolResultEvents.some((e) => e.callId === 'call_pre_retry'),
).toBe(false);
});Reverse-validated: with 合入前完整跑了一遍验证，全包测试 5896 个用例 + typecheck/build 都干净，没有回归。提一个补测试的建议： 现有的 如果没有 修复本身很扎实，把截断检查前移到 schema 校验之前是对的。 |
TaimoorSiddiquiOfficial
pushed a commit
to TaimoorSiddiquiOfficial/HopCode
that referenced
this pull request
Apr 23, 2026
Co-authored-by: Copilot <223556219+Copilot@users.noreply.github.com>
chiga0
pushed a commit
that referenced
this pull request
Apr 24, 2026
* fix(core): reject truncated subagent write_file calls Propagate MAX_TOKENS truncation from subagent responses into tool requests and reject truncated edit calls before schema validation can surface misleading missing-parameter errors. * fix(core): reset per-attempt stream state on retry in agent-core When a streaming response hits MAX_TOKENS and is retried, accumulated state variables (functionCalls, wasOutputTruncated, roundText, etc.) were not cleared. This caused a successful retry to inherit the stale wasOutputTruncated=true flag from the failed attempt, incorrectly rejecting all Edit tool calls with a truncation error. Reset all per-attempt state on retry events, matching the existing behaviour in turn.ts. Add a regression test covering the truncated-then-retried-successfully scenario. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) --------- Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
2 tasks
mabry1985
referenced
this pull request
in protoLabsAI/protoCLI
Apr 26, 2026
…3505, #3467) (#113) * fix: backport upstream trivial bug fixes (#3499, #3630, #3320) Cherry-picked from QwenLM/qwen-code: - #3499 fix(core): use empty string instead of null for reasoning-only assistant content. Some OpenAI-compatible providers (e.g. Ollama qwen3.5:9b) reject content: null with HTTP 400 when reasoning_content is also present. Tool-call-only messages keep null per OpenAI spec. - #3630 fix(telemetry): switch FileExporter.serialize from JSON.stringify to safeJsonStringify. OTel ReadableSpans hold a BatchSpanProcessor back-reference that forms a cycle and crashed --telemetry-outfile users. - #3320 fix(core): cap chokidar depth at 2 in SkillManager and skip .git / special file types. Prevents FD exhaustion when a skill dir contains node_modules etc., which silently broke node-pty I/O. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(core): scope StreamingToolCallParser per-stream (#3525) Backports upstream PR #3525 + extends the per-stream context to also cover our fork's <think>-tag parser state. Bug: every caller of Config.getContentGenerator() — foreground turns, fork subagents, run_in_background subagents, ACP concurrent Agent calls (after #3463) — shared a single OpenAIContentConverter, which held the StreamingToolCallParser as an instance field. Concurrent streams corrupted each other's tool-call buffers, surfacing as NO_RESPONSE_TEXT. Fix: - New ConverterStreamContext interface holds toolCallParser, thinkBuffer, inThinkTag — one per stream. - createStreamContext() factory replaces resetStreamingToolCalls(). - convertOpenAIChunkToGemini(chunk, ctx) and processThinkChunk(chunk, ctx) thread the context through every parser/think-buffer access. - ContentGenerationPipeline.processStreamWithLogging creates one context at stream entry. The error path no longer manually resets — the context is GC'd when the generator unwinds. Our protoInternal recovery-note logic is preserved on the new shape. Note: upstream's follow-up #3550 (full stateless converter refactor) is deferred — it's hygiene without a functional bug; #3525 alone fixes the concurrency race. Tests: - New createStreamContext describe replaces resetStreamingToolCalls suite - Streaming <think> tests use a per-test context - pipeline.test.ts mock updated to match the new API - pipeline.concurrent.test.ts (from upstream commit 38edd9dc5) drives two real concurrent streams and asserts neither corrupts the other's tool-call output (positive control: pre-fix, this test fails with exactly the user-reported bug shape). Refs upstream #3516, #3525. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(core): strip thinking blocks from history on model switch (#3304) When switching models mid-session, reasoning_content fields from thinking-capable models leaked into API requests sent to the new provider, causing 422 errors on strict OpenAI-compatible endpoints. Call stripThoughtsFromHistory() in handleModelChange() so thought parts are removed before the next request is built for the new model. * fix(core): reject truncated subagent write_file calls (#3505) Backport of upstream #3505. Propagates MAX_TOKENS truncation from subagent responses into tool requests and rejects truncated edit calls before schema validation can surface misleading missing-parameter errors. Adapted to our fork's coreToolScheduler.ts which already had the truncation rejection block — kept both, dropped the unused clearRetryCountsForTool() call (we don't have that retry-counter machinery yet). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(core): prevent malformed permission rules from becoming tool-wide catch-alls (#3467) Backport of upstream #3467. A permission rule with unbalanced parens was silently parsed with specifier: undefined, causing matchesRule to treat it as a catch-all. For deny rules this blocked all commands; for allow rules a typo could silently auto-approve everything. - Adds an invalid flag to PermissionRule - parseRule marks unbalanced-paren rules as invalid - matchesRule short-circuits invalid rules to never match - parseRules / addSession*Rule / addPersistentRule warn on malformed input - listRules filters invalid rules from /permissions UI Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(core): preserve reasoning_content during session resume and active sessions (GH#3579) * test(config): drop fork-incompatible #3304 strip-thoughts test The test from upstream #3304 backport assumed an in-place qwen-oauth model switch path that our fork doesn't have; the source-side fix in config.ts (stripThoughtsFromHistory call in handleModelChange) is preserved. Coverage will be re-added when the fork's switch flow stabilizes. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> --------- Co-authored-by: Automaker <automaker@localhost> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com> Co-authored-by: tanzhenxin <tanzhenxing1987@gmail.com> Co-authored-by: Yuchen Fu <fuyuchen0904@163.com>
mabry1985
referenced
this pull request
in protoLabsAI/protoCLI
Apr 26, 2026
#115) * fix: backport upstream trivial bug fixes (#3499, #3630, #3320) Cherry-picked from QwenLM/qwen-code: - #3499 fix(core): use empty string instead of null for reasoning-only assistant content. Some OpenAI-compatible providers (e.g. Ollama qwen3.5:9b) reject content: null with HTTP 400 when reasoning_content is also present. Tool-call-only messages keep null per OpenAI spec. - #3630 fix(telemetry): switch FileExporter.serialize from JSON.stringify to safeJsonStringify. OTel ReadableSpans hold a BatchSpanProcessor back-reference that forms a cycle and crashed --telemetry-outfile users. - #3320 fix(core): cap chokidar depth at 2 in SkillManager and skip .git / special file types. Prevents FD exhaustion when a skill dir contains node_modules etc., which silently broke node-pty I/O. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(core): scope StreamingToolCallParser per-stream (#3525) Backports upstream PR #3525 + extends the per-stream context to also cover our fork's <think>-tag parser state. Bug: every caller of Config.getContentGenerator() — foreground turns, fork subagents, run_in_background subagents, ACP concurrent Agent calls (after #3463) — shared a single OpenAIContentConverter, which held the StreamingToolCallParser as an instance field. Concurrent streams corrupted each other's tool-call buffers, surfacing as NO_RESPONSE_TEXT. Fix: - New ConverterStreamContext interface holds toolCallParser, thinkBuffer, inThinkTag — one per stream. - createStreamContext() factory replaces resetStreamingToolCalls(). - convertOpenAIChunkToGemini(chunk, ctx) and processThinkChunk(chunk, ctx) thread the context through every parser/think-buffer access. - ContentGenerationPipeline.processStreamWithLogging creates one context at stream entry. The error path no longer manually resets — the context is GC'd when the generator unwinds. Our protoInternal recovery-note logic is preserved on the new shape. Note: upstream's follow-up #3550 (full stateless converter refactor) is deferred — it's hygiene without a functional bug; #3525 alone fixes the concurrency race. Tests: - New createStreamContext describe replaces resetStreamingToolCalls suite - Streaming <think> tests use a per-test context - pipeline.test.ts mock updated to match the new API - pipeline.concurrent.test.ts (from upstream commit 38edd9dc5) drives two real concurrent streams and asserts neither corrupts the other's tool-call output (positive control: pre-fix, this test fails with exactly the user-reported bug shape). Refs upstream #3516, #3525. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(core): strip thinking blocks from history on model switch (#3304) When switching models mid-session, reasoning_content fields from thinking-capable models leaked into API requests sent to the new provider, causing 422 errors on strict OpenAI-compatible endpoints. Call stripThoughtsFromHistory() in handleModelChange() so thought parts are removed before the next request is built for the new model. * fix(core): reject truncated subagent write_file calls (#3505) Backport of upstream #3505. Propagates MAX_TOKENS truncation from subagent responses into tool requests and rejects truncated edit calls before schema validation can surface misleading missing-parameter errors. Adapted to our fork's coreToolScheduler.ts which already had the truncation rejection block — kept both, dropped the unused clearRetryCountsForTool() call (we don't have that retry-counter machinery yet). Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(core): prevent malformed permission rules from becoming tool-wide catch-alls (#3467) Backport of upstream #3467. A permission rule with unbalanced parens was silently parsed with specifier: undefined, causing matchesRule to treat it as a catch-all. For deny rules this blocked all commands; for allow rules a typo could silently auto-approve everything. - Adds an invalid flag to PermissionRule - parseRule marks unbalanced-paren rules as invalid - matchesRule short-circuits invalid rules to never match - parseRules / addSession*Rule / addPersistentRule warn on malformed input - listRules filters invalid rules from /permissions UI Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(core): preserve reasoning_content during session resume and active sessions (GH#3579) * test(config): drop fork-incompatible #3304 strip-thoughts test The test from upstream #3304 backport assumed an in-place qwen-oauth model switch path that our fork doesn't have; the source-side fix in config.ts (stripThoughtsFromHistory call in handleModelChange) is preserved. Coverage will be re-added when the fork's switch flow stabilizes. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(acp): run Agent tool calls concurrently + graceful degrade (#3463) Backport of upstream #3463. When the model returns multiple Agent tool calls in a single turn, ACP Session was executing them sequentially in a for-loop, multiplying latency by sub-agent count. - Add private runToolCalls() helper that mirrors coreToolScheduler's partition logic: consecutive Agent calls form a parallel batch (safe because sub-agents have no shared mutable state); other tools form sequential batches. - Replace 2 for-loops in Session.ts with runToolCalls() calls. - Switch the AgentTool eventEmitter guard from key-presence check to truthy check (commit 651979cdf) — the key-presence check passed for { eventEmitter: undefined } and crashed inside SubAgentTracker.setup. Note: upstream replaced 3 for-loops; our fork only had 2 in those code paths. Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com> * fix(acp): support SSE and HTTP MCP servers in ACP mode In ACP mode, the Mcp server list sent by the IDE client can include SSE (type: "sse") and HTTP (type: "http") transports, but the previous implementation only handled stdio servers via toStdioServer(). Non-stdio servers were silently skipped (continue), so any SSE/HTTP-configured MCP server would never be registered. Changes: - Add toSseServer() helper: detects type=="sse" servers and maps them to MCPServerConfig(url=..., headers=...) - Add toHttpServer() helper: detects type=="http" servers and maps them to MCPServerConfig(httpUrl=..., headers=...) - Refactor newSessionConfig() loop to handle all three transport types - Declare mcpCapabilities: { sse: true, http: true } in agentCapabilities so IDE clients know this agent supports these transports without needing a transparent proxy - Export the three helper functions for unit testing Tests: - Unit tests for toStdioServer / toSseServer / toHttpServer helpers (type discrimination, mutual exclusion) - Integration-style tests for QwenAgent.initialize() mcpCapabilities - Integration-style tests for newSession() with SSE/HTTP MCP servers, verifying MCPServerConfig is constructed with the correct arguments (url vs httpUrl, headers passthrough, empty-headers → undefined) Fixes #3472 --------- Co-authored-by: Automaker <automaker@localhost> Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com> Co-authored-by: tanzhenxin <tanzhenxing1987@gmail.com> Co-authored-by: Yuchen Fu <fuyuchen0904@163.com> Co-authored-by: LaZzyMan <zeusdream7@gmail.com>
mabry1985
referenced
this pull request
in protoLabsAI/protoCLI
Apr 26, 2026
* fix: backport upstream trivial bug fixes (#3499, #3630, #3320)
Cherry-picked from QwenLM/qwen-code:
- #3499 fix(core): use empty string instead of null for reasoning-only
assistant content. Some OpenAI-compatible providers (e.g. Ollama
qwen3.5:9b) reject content: null with HTTP 400 when reasoning_content is
also present. Tool-call-only messages keep null per OpenAI spec.
- #3630 fix(telemetry): switch FileExporter.serialize from JSON.stringify
to safeJsonStringify. OTel ReadableSpans hold a BatchSpanProcessor
back-reference that forms a cycle and crashed --telemetry-outfile users.
- #3320 fix(core): cap chokidar depth at 2 in SkillManager and skip
.git / special file types. Prevents FD exhaustion when a skill dir
contains node_modules etc., which silently broke node-pty I/O.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(core): scope StreamingToolCallParser per-stream (#3525)
Backports upstream PR #3525 + extends the per-stream context to also
cover our fork's <think>-tag parser state.
Bug: every caller of Config.getContentGenerator() — foreground turns,
fork subagents, run_in_background subagents, ACP concurrent Agent calls
(after #3463) — shared a single OpenAIContentConverter, which held the
StreamingToolCallParser as an instance field. Concurrent streams
corrupted each other's tool-call buffers, surfacing as NO_RESPONSE_TEXT.
Fix:
- New ConverterStreamContext interface holds toolCallParser, thinkBuffer,
inThinkTag — one per stream.
- createStreamContext() factory replaces resetStreamingToolCalls().
- convertOpenAIChunkToGemini(chunk, ctx) and processThinkChunk(chunk, ctx)
thread the context through every parser/think-buffer access.
- ContentGenerationPipeline.processStreamWithLogging creates one context
at stream entry. The error path no longer manually resets — the
context is GC'd when the generator unwinds.
Our protoInternal recovery-note logic is preserved on the new shape.
Note: upstream's follow-up #3550 (full stateless converter refactor) is
deferred — it's hygiene without a functional bug; #3525 alone fixes
the concurrency race.
Tests:
- New createStreamContext describe replaces resetStreamingToolCalls suite
- Streaming <think> tests use a per-test context
- pipeline.test.ts mock updated to match the new API
- pipeline.concurrent.test.ts (from upstream commit 38edd9dc5)
drives two real concurrent streams and asserts neither corrupts the
other's tool-call output (positive control: pre-fix, this test fails
with exactly the user-reported bug shape).
Refs upstream #3516, #3525.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(core): strip thinking blocks from history on model switch (#3304)
When switching models mid-session, reasoning_content fields from
thinking-capable models leaked into API requests sent to the new
provider, causing 422 errors on strict OpenAI-compatible endpoints.
Call stripThoughtsFromHistory() in handleModelChange() so thought parts
are removed before the next request is built for the new model.
* fix(core): reject truncated subagent write_file calls (#3505)
Backport of upstream #3505. Propagates MAX_TOKENS truncation from
subagent responses into tool requests and rejects truncated edit calls
before schema validation can surface misleading missing-parameter errors.
Adapted to our fork's coreToolScheduler.ts which already had the
truncation rejection block — kept both, dropped the unused
clearRetryCountsForTool() call (we don't have that retry-counter
machinery yet).
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(core): prevent malformed permission rules from becoming tool-wide catch-alls (#3467)
Backport of upstream #3467. A permission rule with unbalanced parens
was silently parsed with specifier: undefined, causing matchesRule to
treat it as a catch-all. For deny rules this blocked all commands; for
allow rules a typo could silently auto-approve everything.
- Adds an invalid flag to PermissionRule
- parseRule marks unbalanced-paren rules as invalid
- matchesRule short-circuits invalid rules to never match
- parseRules / addSession*Rule / addPersistentRule warn on malformed input
- listRules filters invalid rules from /permissions UI
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(core): preserve reasoning_content during session resume and active sessions (GH#3579)
* test(config): drop fork-incompatible #3304 strip-thoughts test
The test from upstream #3304 backport assumed an in-place qwen-oauth
model switch path that our fork doesn't have; the source-side fix in
config.ts (stripThoughtsFromHistory call in handleModelChange) is
preserved. Coverage will be re-added when the fork's switch flow
stabilizes.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(acp): run Agent tool calls concurrently + graceful degrade (#3463)
Backport of upstream #3463. When the model returns multiple Agent tool
calls in a single turn, ACP Session was executing them sequentially in
a for-loop, multiplying latency by sub-agent count.
- Add private runToolCalls() helper that mirrors coreToolScheduler's
partition logic: consecutive Agent calls form a parallel batch (safe
because sub-agents have no shared mutable state); other tools form
sequential batches.
- Replace 2 for-loops in Session.ts with runToolCalls() calls.
- Switch the AgentTool eventEmitter guard from key-presence check to
truthy check (commit 651979cdf) — the key-presence check passed for
{ eventEmitter: undefined } and crashed inside SubAgentTracker.setup.
Note: upstream replaced 3 for-loops; our fork only had 2 in those code
paths.
Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
* fix(acp): support SSE and HTTP MCP servers in ACP mode
In ACP mode, the Mcp server list sent by the IDE client can include
SSE (type: "sse") and HTTP (type: "http") transports, but the previous
implementation only handled stdio servers via toStdioServer(). Non-stdio
servers were silently skipped (continue), so any SSE/HTTP-configured
MCP server would never be registered.
Changes:
- Add toSseServer() helper: detects type=="sse" servers and maps them
to MCPServerConfig(url=..., headers=...)
- Add toHttpServer() helper: detects type=="http" servers and maps them
to MCPServerConfig(httpUrl=..., headers=...)
- Refactor newSessionConfig() loop to handle all three transport types
- Declare mcpCapabilities: { sse: true, http: true } in agentCapabilities
so IDE clients know this agent supports these transports without needing
a transparent proxy
- Export the three helper functions for unit testing
Tests:
- Unit tests for toStdioServer / toSseServer / toHttpServer helpers
(type discrimination, mutual exclusion)
- Integration-style tests for QwenAgent.initialize() mcpCapabilities
- Integration-style tests for newSession() with SSE/HTTP MCP servers,
verifying MCPServerConfig is constructed with the correct arguments
(url vs httpUrl, headers passthrough, empty-headers → undefined)
Fixes #3472
* fix(openai): when samplingParams is set, pass it through verbatim
Previously pipeline.ts always hardcoded max_tokens as the output-token parameter name on the OpenAI-compatible path, falling back from samplingParams.max_tokens to request.config.maxOutputTokens to provider defaults. This broke GPT-5 / o-series on OpenAI and Azure OpenAI, which require max_completion_tokens and reject max_tokens with a 400 error.
Fix: when the user provides samplingParams explicitly, treat it as the complete source of truth for the wire shape and pass its keys through verbatim. No client-injected defaults, no request fallbacks, no hardcoded parameter names. The user describes what the provider wants; the client trusts them.
When samplingParams is absent, the historical default behavior (request fallback through temperature/top_p/.../max_tokens plus provider defaults) is preserved unchanged — existing users see no difference.
Concretely, users can now set any of:
samplingParams: { max_tokens: 4096 } # GPT-4 / Qwen / DeepSeek
samplingParams: { max_completion_tokens: 4096 } # GPT-5 / o-series
samplingParams: { reasoning_effort: 'medium' } # future knobs
without waiting for a qwen-code release that adds model-specific branches.
Signed-off-by: Gordon Lam (SH) <yeelam@microsoft.com>
---------
Signed-off-by: Gordon Lam (SH) <yeelam@microsoft.com>
Co-authored-by: Automaker <automaker@localhost>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>
Co-authored-by: tanzhenxin <tanzhenxing1987@gmail.com>
Co-authored-by: Yuchen Fu <fuyuchen0904@163.com>
Co-authored-by: LaZzyMan <zeusdream7@gmail.com>
Co-authored-by: Gordon Lam (SH) <yeelam@microsoft.com>
xaelistic
pushed a commit
to xaelistic/qwen-code
that referenced
this pull request
Jun 7, 2026
* fix(core): reject truncated subagent write_file calls Propagate MAX_TOKENS truncation from subagent responses into tool requests and reject truncated edit calls before schema validation can surface misleading missing-parameter errors. * fix(core): reset per-attempt stream state on retry in agent-core When a streaming response hits MAX_TOKENS and is retried, accumulated state variables (functionCalls, wasOutputTruncated, roundText, etc.) were not cleared. This caused a successful retry to inherit the stale wasOutputTruncated=true flag from the failed attempt, incorrectly rejecting all Edit tool calls with a truncation error. Reset all per-attempt state on retry events, matching the existing behaviour in turn.ts. Add a regression test covering the truncated-then-retried-successfully scenario. 🤖 Generated with [Qwen Code](https://github.com/QwenLM/qwen-code) --------- Co-authored-by: Qwen-Coder <qwen-coder@alibabacloud.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Fixes #3286
Verification