Skip to content

feat!: remove the kimi-cli backend - #954

Merged
devswha merged 2 commits into
devfrom
bot/remove-kimi-cli
Sep 26, 2026
Merged

devswha merged 2 commits into
devfrom
bot/remove-kimi-cli

Conversation

@devswha

@devswha devswha commented Sep 26, 2026

Copy link
Copy Markdown
Owner

Summary

Removes the kimi-cli backend. Kimi Code takes the prose as a --prompt argv value that other local users can read from the process list, which is why the skill helper already refused it (backend_argv_exposes_input). The Kimi/Moonshot HTTP presets (--provider kimi|moonshot) and the hosted kimi provider stay. Policy: docs/WORKFLOW.md

No Issue: part of the owner-requested 2026-09-26 dead-feature sweep (Tier 2 "kimi-cli backend").

  • Deleted src/backends/kimi-cli.js, scripts/research/kimi-study-transport.mjs, tests/unit/kimi-study-transport.test.js, tests/unit/kimi-isolation.test.js.
  • Registry, selector heuristic (--model kimi-*), batch safety entry, default model, help text, skill-helper guard and SKILL.md sentence, AUTHENTICATION (+KR), CLI.md, FLAG-PARITY.md, .env.example, live-quality help text.
  • Research harness: the kimi transport branches in model-evaluation-transport.mjs, the kimi-only profileIdentityVerified path in study-journal.mjs (acceptedStudyIdentity now takes the call only), and the studySemantics file list.
  • Tests: kimi cases removed from backend/auth/cancellation/model-default/adapter-noise/review roundup #2 (low): preview entity crash, base-href trust, ouroboros/score-gate exit edges, JSON-extractor siblings, +10 #527/auth-login/skill-invocation suites; the fake-CLI isolation capture that only Kimi used is gone.

Scope and risk

  • Contract impact: --backend kimi-cli now exits 2 (Unknown backend: kimi-cli); --model kimi-* no longer routes to a local CLI and stays on openai-http; patina auth login kimi-cli is gone.
  • Not changed: --provider kimi / --provider moonshot, KIMI_API_KEY / MOONSHOT_API_KEY for those presets, the hosted kimi provider, detection, scoring, prompts.
  • Research: journals from the closed 2026-09-05 Kimi study can no longer be replayed; the study's published results in docs/research are unchanged.
  • Size: 33 files, +52 / −710. scripts/rewrite-ab.mjs keeps a string-only kimi-cli judge-family mapping; the dev-tools PR deletes that file.

Verification

  • Rewritten backend-selection tests (kimi-* stays on the HTTP default, kimi-cli is an unknown backend, the fallback hint lists five backends) failed on the unmodified tree (3 fail) and pass after the change.
  • npm test → 0 (2120 tests: 2119 pass, 1 skipped, 0 fail)
  • npm run lint → 0 (architecture: 109 modules, 357 edges, 0 violations)
  • npm run release:check, npm run check:no-private-assets, npm run dogfood, git diff --check → 0
  • echo … | node bin/patina.js --backend kimi-cli --lang en → exit 2 listing the five remaining backends; --list-backends has no kimi row.
  • Base: origin/dev 9108a63.

Release and rollback

  • Semver impact: removes a backend value. Minor per the 2026-09-26 owner policy (9.1.0), stated in the release CHANGELOG semver rationale.
  • Rollback: revert the squash commit.

Kimi Code takes the prose as a --prompt argv value that other local
users can read from the process list, which is why the skill helper
already refused it. `--backend kimi-cli` is now an unknown backend
(exit 2) and `--model kimi-*` stays on openai-http. The Kimi/Moonshot
HTTP presets and the hosted kimi provider stay; the research harness
drops its kimi transport with the backend.
@vercel

vercel Bot commented Sep 26, 2026 •

Copy link
Copy Markdown
Contributor

The latest updates on your projects. Learn more about Vercel for GitHub.

Project Deployment Actions Updated
patina Ready Ready Preview Sep 26, 2026 8:16am UTC

Request Review

Review follow-up for #954: the English and Korean FAQ still offered a logged-in Kimi CLI as an API-key alternative. Also renames a sample backend error in batch.test.js.
@devswha

devswha commented Sep 26, 2026

Copy link
Copy Markdown
Owner Author

Independent review (read-only, one pass): APPROVE (after one fix round)

Reviewer: separate read-only gate-review agent (did not write the change).

  • Round 1: REQUEST_CHANGES. Blocker: docs/FAQ.md and docs/FAQ_KR.md still offered a logged-in "Kimi CLI" in prose, which the kimi-cli grep missed. Fixed in c9b69b0, which also renames the sample backend error in tests/unit/batch.test.js.
  • Round 2 (same reviewer, follow-up commit only): APPROVE. No current-behavior doc mentions a Kimi CLI any more.
  • Non-blocking, kept on purpose: scripts/research/study-journal.mjs still recognises the 'cli-request-trace' / 'cli-session-trace' evidence values, so the journals of the closed 2026-09-05 Kimi study keep being read faithfully. The rewrite-ab kimi mapping is gone with chore: remove retired and duplicate dev tooling #955.

Verified by the reviewer: HTTP kimi/moonshot presets and the hosted kimi provider untouched; claude/codex/gemini/agy routing, caps, and defaults unchanged; research harness semantics identical for the remaining transports; 14 targeted test files 175/175; check-architecture 0 violations; --backend kimi-cli exits 2 listing five backends.

@devswha
devswha marked this pull request as ready for review September 26, 2026 08:18
@devswha
devswha merged commit 7f52475 into dev Sep 26, 2026
8 checks passed
@devswha
devswha deleted the bot/remove-kimi-cli branch September 26, 2026 08:18

This branch was successfully deployed

1 active deployment
Preview — c9b69b0c Deployed Sep 26, 2026 by vercel[bot]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant