fix: mimo-parallel-tool-call-policy (2/2) by myk1yt · Pull Request #32 · myk1yt/Zoo-Code

myk1yt · 2026-08-02T05:39:02Z

Stack Position

Feature Branch: fix/mimo-parallel-tool-call-policy
Stage: 2/2
Depends on: feat: openai-compatible-strict-reasoning (1/2) #26

Description

Full Feature Description

Feature Branch: fix/mimo-parallel-tool-call-policy
Feature Name: MiMo Parallel Tool Call Policy
Purpose: Resolves the problem where MiMo's OpenAI-compatible stream can send two or more tool calls, ghost calls with no name or arguments, and incomplete JSON arguments even in situations where only a single call should be safely processed. Prevents ambiguous side-effect ordering, executes only structurally valid calls, and preserves the existing parallel behavior of other providers.
Full Change Description: The shared root B05a provides OpenAI-compatible schema conversion and provider-neutral reasoning/strict controls. B12 connects MiMo model capability, request-level parallel control, streamed native call parsing, ghost quarantine, max-one retention policy, task execution guard, and privacy-safe telemetry. Under the single-call policy, if there is exactly one valid call, only that call is executed; if there are two or more, none are auto-selected and all are returned as typed error results so the model resubmits only one.
Impact Scope: Affects mimo.ts, mimo.ts, NativeToolCallParser.ts, ToolCallRetentionPolicy.ts, task policy wiring, and TelemetryService.ts. B05a is shared between this chain and the openai-compatible-strict-reasoning chain.
Errors and Edge Cases: Only terminated ghosts with neither name nor arguments are silently removed. If name or argument bytes are present, malformed calls are not erased from history and produce exactly one error result. When two or more valid side-effect calls arrive, none are auto-executed. Telemetry records only counts and policy metadata, not tool names, call IDs, arguments, commands, paths, or prompts. Known parallel provider behavior for OpenAI/Anthropic families is not regressed.
Testing Method: Run B05a's schema conversion tests and B12's MiMo single/parallel/malformed/ghost stream fixture, native parser, retention policy, telemetry, and task policy tests. Manually replay a MiMo response with two tool calls, verifying that zero executions and two error results are produced, that a single call is executed exactly once, and that ghosts are removed before history.

Why Split Into 17 PRs

Instead of submitting this feature as a single unified PR, it was split into individual PRs because as code size grows, safely reviewing a PR becomes very difficult. The feature was broken into mutually exclusive individual PRs so that each can be reviewed independently.

What This PR Specifically Changes

Adds MiMo capability/request conversion, stream parser, ghost quarantine, malformed call retention, max-one execution guard, and payload-free telemetry. Does not change other provider behavior or cost calculation.

Included Files

packages/types/src/providers/mimo.ts
packages/telemetry/src/TelemetryService.ts
src/api/providers/mimo.ts
src/core/assistant-message/NativeToolCallParser.ts
src/core/assistant-message/ToolCallRetentionPolicy.ts
task policy wiring and direct tests

Exclusion Scope

Request/cost changes for other providers
Duplicate changes to B05a settings/UI/request shaping
Session report and repair script
Changes that log prompt or tool argument payloads to telemetry
All items in the common removal rules

# Conflicts: # src/core/tools/error-interception/StructuralValidator.ts

# Conflicts: # src/core/assistant-message/__tests__/NativeToolCallParser.spec.ts # src/core/assistant-message/__tests__/presentAssistantMessage-parser-dedup.integration.spec.ts # src/core/assistant-message/presentAssistantMessage.ts

# Conflicts: # src/core/assistant-message/__tests__/presentAssistantMessage-parser-dedup.integration.spec.ts

…capabilities

…ception refs from backup

MimoHandler was passing raw tool schemas to the API without the strict mode conversion that all other OpenAI-compatible providers use. This caused tool call errors due to missing required/strict fields. - Call this.convertToolsForOpenAI(tools) instead of raw assignment - Adds strict: true, required properties, additionalProperties: false

…de arg

An id-less argument-continuation chunk belongs to the most recent id chunk seen at its index. When a provider reuses index 0 with a NEW id (a disguised second parallel call), the new call's id chunk was dropped but its id-less argument fragments were still kept and concatenated into the FIRST call's accumulator, corrupting its JSON. Track dropped indexes in filterToFirstToolCall state and drop subsequent id-less fragments for those indexes. Also rewrite the function docblock, which referenced a non-existent error-interception retry loop.

The parseErrors/parseFailures docblocks claimed presentAssistantMessage routes recorded failures to an INVALID_JSON_ARGUMENTS error-interception pattern. No such routing exists on this codebase; describe the actual lifecycle (consumed via the consume* APIs, cleared on new API request). Comment-only change, no behavior difference.

parseErrors/parseFailures static maps accumulated an entry per malformed tool call and were never cleared in production (the consume* APIs have no production callers), slowly leaking for the extension-host lifetime. Add NativeToolCallParser.clearParseFailures() and call it in Task.recursivelyMakeClineRequests alongside clearAllStreamingToolCalls()/ clearRawChunkState(), where other per-stream state is reset. The consume* APIs keep working for tests.

MiMo sends tools through convertToolsForOpenAI(), which attaches a strict flag to every function tool. An OpenAI-compatible endpoint that doesn't support structured outputs rejects the request with a 400 and the turn fails outright. Mirror the existing parallel_tool_calls fallback: detect schema-rejection errors narrowly (400 status plus a mention of strict/additionalProperties in a tools context, so unrelated 400s like MiMo's missing-reasoning_content rejection are not retried) and retry once with the original schemas and no strict flag.

…d telemetry events

…oo-Code-Org#1130) CI failures: 1. Code QA Roo Code run 31229619240: unused 'taskId' binding (lint). 2. E2E Tests (Mocked) run 31229619233: 4 TS errors in mimo-parallel.test.ts: - mimoBaseUrl was a 4-literal union, rejecting the local mock server URL - 'enableToolUse' is not a RooCodeSettings key - 'api_req_failed' is a ClineAsk, not a ClineSay (x2) Fix: - provider-settings.ts: widen mimoBaseUrl to z.string().url() (documented with the 4 common endpoints) so tests can point the handler at a local mock. - mimo-parallel.test.ts: drop unused taskId, remove enableToolUse, compare m.ask === 'api_req_failed' instead of m.say. Runs: https://github.com/Zoo-Code-Org/Zoo-Code/actions/runs/31229619240 https://github.com/Zoo-Code-Org/Zoo-Code/actions/runs/31229619233

…parallel test

…o-parallel test

Co-authored-by: Roomote <roomote@roomote.dev>

…sages load (Zoo-Code-Org#1181) * fix(task): skip saveClineMessages when history task aborts before messages load * test: strengthen resume-eviction-race assertions and type mock provider * test: add fallback mock for second getSavedClineMessages read in resume-eviction spec * fix(task): prevent saving unhydrated history messages during abort --------- Co-authored-by: Naved Merchant <naved.merchant@gmail.com>

…-Code-Org#1198) Co-authored-by: Roomote <roomote@roomote.dev>

…rg#1141) * refactor(webview): complete provider identifier migration * test(webview): type selected model hook mocks * test(webview): keep selected model spec out of provider migration * test(webview): harden provider identifier migration tests --------- Co-authored-by: Elliott de Launay <edelauna@gmail.com>

…de-Org#1193) * chore(deps): update dependency mermaid to v11.16.1 [security] * fix(webview): harden mermaid securityLevel and tighten dev-mode CSP --------- Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com> Co-authored-by: Elliott de Launay <edelauna@gmail.com>

…-Org#1161) Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>

…#1199) Co-authored-by: Roomote <roomote@roomote.dev>

…o-Code-Org#1200) Co-authored-by: Roomote <roomote@roomote.dev>

…rg#1146) * refactor(webview): canonicalize ApiOptions provider identifiers * test(webview): strengthen ApiOptions interaction coverage * fix(webview): use providerIdentifiers.openrouter in ApiOptions option sort --------- Co-authored-by: Elliott de Launay <edelauna@gmail.com>

…-Code-Org#1203) Co-authored-by: Roomote <roomote@roomote.dev>

) Co-authored-by: Roomote <roomote@roomote.dev>

…o-Code-Org#1204) Co-authored-by: Roomote <roomote@roomote.dev>

…-Code-Org#1201) Co-authored-by: Roomote <roomote@roomote.dev>

…t EnvironmentTeardownError

myk1yt · 2026-08-10T11:54:45Z

Closing to recreate with main as target base branch. This PR had stale base branch references after fork sync.

myk1yt changed the title ~~B12: MiMo Enforcement v2~~ fix: mimo-parallel-tool-call-policy (2/2) Aug 3, 2026

myk1yt force-pushed the pr/b05a-strict-reasoning-v2 branch from 658c4f2 to 4692197 Compare August 4, 2026 11:34

myk1yt force-pushed the pr/b12-mimo-enforcement-v2 branch from 1cf9a96 to 0532102 Compare August 4, 2026 11:40

myk1yt force-pushed the pr/b05a-strict-reasoning-v2 branch from 4692197 to 2adc65a Compare August 4, 2026 20:28

myk1yt force-pushed the pr/b12-mimo-enforcement-v2 branch from b526a57 to 132fb08 Compare August 4, 2026 20:28

myk1yt force-pushed the pr/b05a-strict-reasoning-v2 branch 2 times, most recently from 77e7207 to a3b22a7 Compare August 4, 2026 20:40

myk1yt force-pushed the pr/b12-mimo-enforcement-v2 branch 4 times, most recently from 67850e5 to c0e5951 Compare August 6, 2026 04:38

Zoo (VP) and others added 16 commits August 6, 2026 13:50

feat: add model-level tool-call capability and policy resolution

3e60771

feat: wire MiMo provider controls and tighten argument normalization

51fc0ba

# Conflicts: # src/core/tools/error-interception/StructuralValidator.ts

feat: add tool-call policy telemetry events

d56b7fd

# Conflicts: # src/core/assistant-message/__tests__/presentAssistantMessage-parser-dedup.integration.spec.ts

fix: resolve no-explicit-any lint errors in mimo and telemetry files

c54608b

fix: preserve parallel behavior for known providers without explicit …

05341fd

…capabilities

fix: port NativeToolParseFailure infrastructure and clean error-inter…

9fa4256

…ception refs from backup

fix: port cleaned mimo.ts provider from backup to match spec types

d759504

fix(mimo): suppress parallel tool calls at provider stream level

12fb091

fix(mimo): pass openAiToolStrictMode setting to convertToolsForOpenAI

169c842

fix(mimo): pass tools to convertToolsForOpenAI without extra strictMo…

0cd4577

…de arg

myk1yt force-pushed the pr/b12-mimo-enforcement-v2 branch from c0e5951 to 86477b0 Compare August 6, 2026 05:10

Zoo (VP) added 2 commits August 7, 2026 01:10

test(b12): add coverage tests for tool-call-policy streaming state an…

b9c3687

…d telemetry events

ci: retrigger workflow after transient infra outage

ff3662a

chore: remove temporary docs and scripts from PR diff

5c376dc

myk1yt deleted the branch pr/b05a-strict-reasoning-v2 August 7, 2026 13:05

myk1yt closed this Aug 7, 2026

myk1yt deleted the pr/b12-mimo-enforcement-v2 branch August 7, 2026 13:06

myk1yt restored the pr/b12-mimo-enforcement-v2 branch August 7, 2026 13:32

test(e2e): add MIMO parallel enforcement suite

520ccb1

myk1yt reopened this Aug 8, 2026

Zoo (VP) and others added 17 commits August 8, 2026 15:13

fix(vscode-e2e): fix mimo-parallel test base url and type assertions

64d6569

fix(vscode-e2e): capture say messages with partial undefined in mimo-…

8d52e02

…parallel test

fix(vscode-e2e): capture ask tool messages and complete Turn 2 in mim…

b745aef

…o-parallel test

refactor: reuse webview chat render helper (Zoo-Code-Org#1196)

84e54cf

Co-authored-by: Roomote <roomote@roomote.dev>

refactor: reuse shared test helpers in config import/export spec (Zoo…

ca86e38

…-Code-Org#1198) Co-authored-by: Roomote <roomote@roomote.dev>

chore(deps): update dependency undici to v6.28.0 [security] (Zoo-Code…

2b2641b

…-Org#1161) Co-authored-by: renovate[bot] <29139614+renovate[bot]@users.noreply.github.com>

refactor: reuse shared reset helper in code-index specs (Zoo-Code-Org…

2a62b64

…#1199) Co-authored-by: Roomote <roomote@roomote.dev>

refactor: reuse shared reset helper in remaining code-index specs (Zo…

25fa8d4

…o-Code-Org#1200) Co-authored-by: Roomote <roomote@roomote.dev>

refactor: reuse webview render helper in chat and settings specs (Zoo…

02f7902

…-Code-Org#1203) Co-authored-by: Roomote <roomote@roomote.dev>

refactor: reuse shared reset helper in provider specs (Zoo-Code-Org#1202

20dd966

) Co-authored-by: Roomote <roomote@roomote.dev>

refactor: reuse webview render helper in remaining settings specs (Zo…

03db526

…o-Code-Org#1204) Co-authored-by: Roomote <roomote@roomote.dev>

refactor: reuse shared reset helper in semble and terminal specs (Zoo…

4c42f91

…-Code-Org#1201) Co-authored-by: Roomote <roomote@roomote.dev>

myk1yt force-pushed the pr/b12-mimo-enforcement-v2 branch from 4ca9b06 to b745aef Compare August 9, 2026 06:30

myk1yt and others added 3 commits August 9, 2026 15:30

Merge branch 'main' into pr/b12-mimo-enforcement-v2

849457d

fix: add missing clearAllMocks import in mimo.spec.ts

e935251

fix(test): add deterministic mocks to Task.throttle.test.ts to preven…

055b3b3

…t EnvironmentTeardownError

myk1yt closed this Aug 10, 2026

Navigation Menu

fix: mimo-parallel-tool-call-policy (2/2) - #32

fix: mimo-parallel-tool-call-policy (2/2)#32
myk1yt wants to merge 46 commits into
pr/b05a-strict-reasoning-v2from
pr/b12-mimo-enforcement-v2

myk1yt commented Aug 2, 2026 •

edited

Loading

Uh oh!

myk1yt commented Aug 10, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Conversation

myk1yt commented Aug 2, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Stack Position

Description

Full Feature Description

Why Split Into 17 PRs

What This PR Specifically Changes

Included Files

Exclusion Scope

Uh oh!

myk1yt commented Aug 10, 2026

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

myk1yt commented Aug 2, 2026 •

edited

Loading