docs: consolidate Codex, Grok, and OpenCode runtime status - #443
Conversation
|
分诊(2026-07-29,通信龙):建议保留,小幅更新后可合。 内容质量高,尤其是这几点符合我们的既有标准:
今天早上 Vincent 直接问过「opencode 做得怎么样了」「codex TUI 共存进 npm 了吗」—— 本文正是答案。它躺在草稿里没人合,导致这个问题被重新查了一遍。 需要更新的(版本号漂移):文中引用 补充今天的实测供更新参考:
状态 MERGEABLE / CLEAN,无冲突。 |
|
分流结论:面向用户的那份已经落地在更好的位置,建议关闭。 本 PR 想加 它已经在 anet.sh 上线,读者能直接看到;仓内再放一份等价内容会形成两个真相源, 如果本 PR 里有 按「不能确立就不关」原则,我没关 —— 我无法替你判断那部分差异是否重要。 |
|
Closing as an obsolete status snapshot; preserving the branch and commit history. This draft is dated 2026-07-13 and makes time-sensitive current-state claims that are contradicted by present
Because this file presents itself as a consolidated current status page rather than an immutable dated test report, merging it now would reintroduce known-false operational guidance. Any useful historical evidence remains reachable in PR #443/commit No branch deletion, force-push, code change, package release, deployment, or production action was performed. |
What
/and mark the RFC-030: wire codex-app-server into anet CLI #441 CLI command as a draft-branch candidate, not a published guaranteeopencode-cliACP runtime from MCP-only human-driven accessWhy
The existing runtime evidence lives across RFCs, issues, draft PRs, fixtures, and research reports. This page consolidates it without promoting a fixture, source path, or internal freeze into a release claim.
Impact
Docs only. This PR does not unlock RFC-030 §8, Wave 2, merge, deployment, production, package promotion, or
latest. It also does not unlock Grok Phase 1A or claim that OpenCode CLI onboarding is complete.Checks
git diff --cached --checkorigin/mainnpm view