feat(chat): expose Kimi ACP config controls - #1261
Conversation
|
The latest updates on your projects. Learn more about Vercel for GitHub. |
Bugbot couldn't run - usage limit reachedBugbot is counted against Cursor usage for this user or team, and this run hit a usage or spend limit. A user or team admin can review and increase usage limits in the Cursor dashboard. (requestId: serverGenReqId_f54978b3-dfa3-455f-be56-fbb3bb6c4f7b) |
|
No actionable comments were generated in the recent review. 🎉 ℹ️ Recent review info⚙️ Run configurationConfiguration used: Repository: arul28/ADE/.coderabbit.yaml Review profile: CHILL Plan: Advanced Run ID: ⛔ Files ignored due to path filters (2)
📒 Files selected for processing (7)
Included review availability: Your plan provides up to 2 included reviews per hour; 0 remain after this review. 📝 WalkthroughWalkthroughChangesKimi ACP configuration
Priority: ⬇️ Low Estimated code review effort: 2 (Simple) | ~10 minutes Change: Feature Suggested labels: Merge Risk: ⚪ Minimal · up to The Kimi configuration flow is covered by contract tests and does not introduce a verified runtime regression. 🚥 Pre-merge checks | ✅ 4 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (4 passed)
Full details: Docstring CoverageExplanation Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%. Docstring coverage is scoped to functions touched by this diff. Analyzed 1 functions across 5 files. (2 skipped: 2 too large.)
✨ Finishing Touches 💡 1📝 Generate docstrings 💡
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
![devin-ai-integration[bot]](https://avatars.githubusercontent.com/in/811515?s=60&v=4){kind=small}
There was a problem hiding this comment.
Devin Review found 1 potential issue.
2 flags not posted on this PR by your GitHub settings — view them in Devin Review. (Configure)
| mcpInjection: capability(transportGatedMcpInjection), | ||
| imagePrompts: capability(inlineImagePrompt), | ||
| configOptionIds: [], | ||
| configOptionIds: KIMI_CONFIG_OPTION_IDS, |
There was a problem hiding this comment.
🟡 Kimi thinking control remains unavailable
KIMI_CONFIG_OPTION_IDS declares thinking, but no production consumer reads configOptionIds or acpConfigSnapshot. Kimi users cannot view or change the advertised thinking option.
Learn more
The dialect list is only metadata. Its current references are tests, while acpConfigSnapshot is produced and persisted by applyAcpConfigOptions but has no renderer consumer. Existing ACP permission and model controls use separate session fields, so they do not render the returned thinking option.
Example: Kimi 2.0 returns a thinking select with off, low, and high. ADE stores that option in the session snapshot, but the composer renders no matching selector, leaving the user unable to choose high.
Recommended fix: Thread acpConfigSnapshot into the composer control model and add an update contract for non-mode ACP options. Render only options returned by the live agent and permitted by dialect.configOptionIds, then route changes to the active session's setConfigOption behavior and persist the resulting snapshot.
Was this helpful? React with 👍 or 👎 to provide feedback.
1 participant
Problem
Kimi Code 2.0.0 documents ACP session configuration for mode, model, and thinking, but ADE marked Kimi session config absent and hid those controls.
Cause
The Kimi dialect was still modeled only from the captured 0.39.1 lifecycle fixture.
Change and boundary
Enable the existing standard
session/set_config_optionACP behavior for Kimi and declare the three vendor-documented option IDs. Keep the existing pooled lifecycle, hidden usage meter, model launch mapping, and provider limitations unchanged. Update the ACP docs and contract tests.Verification
Focused ACP and chat-runtime tests: 1,293 passed. Desktop typecheck passed. A live unauthenticated Kimi 0.39.1 handshake passed. Affected CI-shaped desktop shards 3/8, 4/8, and 8/8 passed (7,548 tests passed; 15 expected skips). Docs validation passed for 265 files.
Model/harness: GPT-5 Codex via ADE Work chat.
Note
Low Risk
Dialect and test/doc updates only; behavior aligns Kimi with existing Qwen session-config plumbing without changing auth, pooling, or usage handling.
Overview
Kimi ACP now advertises
session/set_config_optionformode,model, andthinking, matching Kimi Code 2.0.0 docs. The dialect switches from absent session config tostandardSetConfigOption, exportsKIMI_CONFIG_OPTION_IDS, and keeps spawn argv, pooledsession/close, and hidden usage unchanged.Runtime: In yolo, Kimi is grouped with Qwen—permission posture goes through the agent via config options, so ADE does not client-side auto-approve permission prompts when
sessionConfigis declared.Tests & docs: Fixture and host tests assert Kimi forwards config RPCs; Kimi is removed from the “refuses config option” provider list. ACP provider spec and verification brief document 2.0.0 config controls and the 0.39.1 baseline.
Reviewed by Cursor Bugbot for commit bb964f4. Configure here.
Summary by CodeRabbit
New Features
Tests
Documentation