Release 0.2.1: docs polish, K3 rationale, real-run screenshots, version bump - #3
Merged
Merged
Conversation
Task lifecycle / session safety: - --timeout keeps the ACP session resumable: on request timeout the client sends session/cancel, failed jobs record sessionId, and --resume reuses the exact same thread (verified end-to-end with real kimi). - Cancel race (POSIX): failOrphan skips jobs already marked cancelled. Acceptance fixes: - Plan mode with read-only tools but zero plan text is now a failed handoff (ok:false, exit 1, planEmptyText:true) — the plan itself is the deliverable in plan mode; previously it passed with ok:true and empty text. - Goal framing no longer bypasses Q&A / reply-exactly exclusion: --goal with a "Reply with exactly: …" or how-to Objective no longer wastes Mode B continue nudges (previously asGoal returned action unconditionally). Hardening: - Mode A empty-turn retries default 2→5, configurable via --empty-retries or KIMI_EMPTY_RETRIES (0 disables). - task/goal reject unknown --flags instead of leaking them into Kimi's prompt. - Media TOCTOU (existsSync→statSync race) is a clean error, not a raw throw. - Job pruning removes logs/<job>.log with the job file; prunes throttled 30s. - status --wait / result --wait exit non-zero while the job is still running. - splitRawArgumentString handles escaped quotes/backslashes inside quotes. - Docs: default mode documented as full-auto approval (plan rejects writes); command docs + AGENTS.md + CHANGELOG updated for 0.2.1. Tests (90, kimi-free): - fake-kimi ACP fixture; acp-client (timeout session retention, empty-retry budget); companion-cli (wait exit codes, orphan reconcile, cancel race, resume hint, unknown flag); turn-policy (plan acceptance, goal framing); host-free-attack (planEmptyText contract); args; state.
…on bump to 0.2.1 - Bump all version fields 0.2.0 -> 0.2.1 (package.json, marketplace, plugin.json, companion, acp-client, smoke, README x2, AGENTS) - README (CN+EN): add K3 benchmark rationale (nxcode + Hugging Face sources), auto-trigger explanation, real-run screenshot (LIMS to Linear light, 2m55s) - README compatibility table: pin minimum Kimi Code >= 0.30.0 - SECURITY.md: fix Supported versions (0.4.x -> 0.2.x) - bug_report.yml placeholder: 0.1.0 -> 0.2.1 - CHANGELOG: mark 0.2.1 as released (drop in-development tag) - Add docs/assets/cc_test.png
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
# Conflicts: # CHANGELOG.md # CHANGELOG.zh-CN.md
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
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
Companion hardening (#2) + documentation/release polish for 0.2.1.
Documentation & marketing
Bug fixes
Version bump (0.2.0 → 0.2.1)
All version fields synced:
package.json,marketplace.json,plugin.json, companionVERSION,acp-clientPLUGIN_VERSION,smoke.mjs, README badges/tables,AGENTS.md.CHANGELOG
in developmenttag)Checklist
npm testpasses (90 tests)