diff --git a/README.ja-JP.md b/README.ja-JP.md index 105dc09..0c705b7 100644 --- a/README.ja-JP.md +++ b/README.ja-JP.md @@ -123,8 +123,8 @@ DeepSeekCode は自身の開発に使える段階ですが、Claude Code CLI / C ほどの製品成熟度にはまだ届いていません。Linux/macOS のローカル coding-agent CLI に絞ると、残差は主に evidence depth と配布面の polish です。 -- macOS shell/runtime evidence を entrypoint smoke 以上に厚くすること。CI/release - matrix は `agents shell-fixture-smoke` と `agents service-smoke` を実行します。 +- macOS shell/runtime の hosted CI evidence は PR #14 / CI run #35 で通過済み。 + release binary evidence は次回 release matrix run で取得します。 - よりリッチな multi-file external fixture サンプル。disposable Python invoice fixture の scaffold script を追加しています。 - Homebrew 公開。tap 資格情報が未設定です。 diff --git a/README.md b/README.md index b860c77..be029f9 100644 --- a/README.md +++ b/README.md @@ -122,9 +122,9 @@ DeepSeekCode is close enough to use as its own coding CLI, but it is not yet at Claude Code CLI / Codex CLI polish. For a Linux/macOS local coding-agent CLI, the remaining gaps are mostly evidence depth and distribution polish: -- macOS shell/runtime CI and release-binary evidence beyond the entrypoint - smoke, now covered by `agents shell-fixture-smoke` and `agents service-smoke` - workflow gates; +- macOS shell/runtime CI evidence beyond the entrypoint smoke is now recorded + in PR #14 / CI run #35; release-binary evidence will come from the next + release matrix run; - richer multi-file external fixture samples, with a scaffold script for a disposable Python invoice fixture; - Homebrew publishing, still blocked on tap credentials; diff --git a/README.zh-CN.md b/README.zh-CN.md index c60d370..b6d5606 100644 --- a/README.zh-CN.md +++ b/README.zh-CN.md @@ -118,8 +118,8 @@ DeepSeekCode 已经可以直接拿来写自己的代码,但还没有达到 Cla Codex CLI 的产品成熟度。如果只看 Linux/macOS 本地 coding-agent CLI,剩余差距主要是 证据厚度和分发打磨: -- macOS shell/runtime 证据需要超过入口 smoke;CI/release matrix 现在会跑 - `agents shell-fixture-smoke` 和 `agents service-smoke`; +- macOS shell/runtime 的 hosted CI 证据已经在 PR #14 / CI run #35 通过; + release binary 证据等待下一次 release matrix 产出; - 更丰富的 multi-file external fixture 样本;仓库已提供 disposable Python invoice fixture 脚手架; - Homebrew 发布仍缺 tap 凭据; diff --git a/docs/current-status.md b/docs/current-status.md index 411ad98..0e73429 100644 --- a/docs/current-status.md +++ b/docs/current-status.md @@ -41,6 +41,9 @@ review、resume 和本地 runtime/shell-supervisor 工作流,就可以认为 `agents shell-fixture-smoke --json`、`agents service-smoke --json`、TUI entrypoint smoke、task worktree smoke 和 GitHub bridge smoke;这把 Linux/macOS 本地 CLI 的入口、 runtime、shell-supervisor 和后台 worktree 基线纳入同一类 release gate。 + PR #14 的 CI run #35 已在 hosted Linux/macOS debug binary 上通过 shell fixture、 + service smoke 和 multi-file external fixture scaffold gate: + https://github.com/willamhou/DeepSeekCode/actions/runs/26333425574 。 - 本轮新增:`deepseek chat` / `deepseek repl` / `deepseek interactive` 的真实 TTY 输入现在走内置 raw-mode line editor,补齐 Claude Code-like REPL 的 Up/Down history、history draft restore、左右移动、Home/End、Backspace/Delete、Ctrl+A/E/U/K/W、Tab slash/session completion、空行 Ctrl+D 和 prompt Ctrl+C 退出;运行中的 REPL turn 也会把 SIGINT 接到 `AgentLoopOptions.cancel_check`,让模型 stream 和 cancel-aware tools 协作取消,并在取消后恢复本轮 transcript/snapshot 指针,避免半截 prompt 污染后续上下文;`/sessions [prefix]` 可以列出保存的 REPL session,`/load ` 后 Tab 可补全 session 名;非交互测试路径仍保留 buffered reader,不需要真实终端。 - 本轮新增:Phase 12E background worktree runner 第一片。新增 `deepseek task start/list/show/stop/diff/merge/reject` 和 `deepseek task fixture-smoke --json`:`task start` 会在当前 git repo 的 `.dscode/task-runner/worktrees/` 创建隔离 worktree 和默认 `deepseek-task/` 分支,把记录写到 `.dscode/task-runner/records/`,stdout/stderr 写到 `.dscode/task-runner/logs/`,并在该 worktree 中启动 `deepseek exec --json`;父 CLI 退出后 child 进程仍可继续。`--no-run` 可只创建 worktree/record,用于无 API key 的本地 gate;`task diff` 展示 task worktree 的 tracked patch/stat 和 untracked files,`task merge --check` dry-run 验证,`task merge` 要求原 worktree 干净后把 patch 和 untracked regular files 合回原 repo,`task reject` 默认删除受管 task worktree 并把记录标记为 rejected。`deepseek github action --background-task` 现在也可把解析出的 GitHub PR review/fix/patch 请求委派到同一 task runner,`--task-id` 支持 workflow 稳定 id,`--task-no-run` 支持无凭据本地 workflow gate。当前 `task fixture-smoke --json` 实测 `ok=true`、`worktree_created=true`、`record_listed=true`、`merge_check_ok=true`、`merge_apply_ok=true`、`reject_ok=true`、`cleanup_ok=true`;CI 已把该 smoke 接到 Linux/macOS/Windows debug binary,Release Matrix 也会在各平台 release binary packaging 前运行。 - 本轮新增:`deepseek agents shell attach --interactive` / `--takeover`。它会进入本地 raw mode,把按键转发到 supervisor `stdin`,把 resize 转发到 supervisor `resize`,并把 output 事件的 raw bytes replay 回当前终端;Linux 集成 smoke 已覆盖 raw-mode PTY 启动、`tty=true` job、stdin、resize、replay 和 bounded detach。它是可用的 bounded interactive attach,不是字节级 PTY fd 直连代理。 @@ -89,9 +92,9 @@ deepseek agents shell-fixture-smoke --json 当前距离 Claude Code CLI / Codex CLI / DeepSeek-TUI 的成熟产品形态,主要差在以下几类: 如果只看 Linux/macOS 本地 CLI milestone,核心交互能力已经成立;剩下主要是 -Homebrew 发布凭据、macOS CI/release smoke 证据落地、更多 online multi-file external -fixture 样本和文档压缩。Windows/IDE/hosted 发布证据继续保留在更大产品目标里,但不是 -这个 milestone 的 blocker。 +Homebrew 发布凭据、下一次 release matrix 的 release-binary smoke 证据、更多 online +multi-file external fixture 样本和文档压缩。Windows/IDE/hosted 发布证据继续保留在更大 +产品目标里,但不是这个 milestone 的 blocker。 1. Shell/PTY 深水区 - 已有 bounded interactive attach、duplex `byte_stream` raw-output proxy slice、human `agents shell proxy` raw-mode wrapper、Windows `native-supervisor` ConPTY backend compile gate,以及 Linux 本地 `pty_fd` / SCM_RIGHTS PTY master fd handoff slice。 @@ -128,8 +131,10 @@ fixture 样本和文档压缩。Windows/IDE/hosted 发布证据继续保留在 1. 固化 Linux/macOS CLI release gate - CI 和 Release Matrix 已新增非 Windows `agents shell-fixture-smoke --json`、 `agents service-smoke --json` 和 multi-file external fixture scaffold smoke。 - - 下一步等 GitHub Actions 跑出 Linux/macOS debug/release 证据后,把对应 run 链接写回 - status/spec;Windows shell-supervisor 继续作为后续跨平台目标。 + - 已记录 PR #14 / CI run #35 的 Linux/macOS debug binary 证据: + https://github.com/willamhou/DeepSeekCode/actions/runs/26333425574 。 + 下一步等待下一次 release matrix 产出 release-binary 证据;Windows + shell-supervisor 继续作为后续跨平台目标。 2. 补外部 model-backed 证据和真实 demo - 先轮换任何已经泄漏到聊天记录里的 key。 @@ -158,10 +163,10 @@ DeepSeekCode 现在已经是一个可以实际使用的 code agent CLI,尤其 但它还不是“可以公开宣称等同 Claude Code CLI / Codex CLI”的成熟产品。 如果目标限定为 Linux/macOS 本地 code agent CLI,则当前判断更强:Linux 本机已经通过 -TUI entrypoint、shell fixture、service smoke 和在线 dogfood release gate;macOS 入口 -smoke 已有,shell/runtime smoke 已加入 CI/Release Matrix 等待 hosted run 产出。Windows -不再影响这个限定目标。 +TUI entrypoint、shell fixture、service smoke 和在线 dogfood release gate;PR #14 / CI +run #35 已在 hosted macOS 上通过 shell/runtime smoke 和 multi-file scaffold gate。 +Windows 不再影响这个限定目标。 最准确的公开表述是: -> DeepSeekCode is usable today for Linux/macOS dogfooding and repository work, with a full-screen TUI, REPL, durable runtime, permissioned tools, shell-supervisor smoke gates, release binaries, a 100-run online dogfood release gate, initial external disposable-repo write-fixture evidence, real hosted GitHub workflow evidence, and a committed real model-backed README demo SVG. The remaining Linux/macOS CLI work is Homebrew publishing, richer online multi-file fixture evidence, and documentation polish; hosted IDE, Windows/service proof, npm publishing, and optional richer demo media remain broader product-hardening work. +> DeepSeekCode is usable today for Linux/macOS dogfooding and repository work, with a full-screen TUI, REPL, durable runtime, permissioned tools, hosted Linux/macOS shell-supervisor smoke gates, release binaries, a 100-run online dogfood release gate, initial external disposable-repo write-fixture evidence, real hosted GitHub workflow evidence, and a committed real model-backed README demo SVG. The remaining Linux/macOS CLI work is Homebrew publishing, richer online multi-file fixture evidence, next-release binary smoke evidence, and documentation polish; hosted IDE, Windows/service proof, npm publishing, and optional richer demo media remain broader product-hardening work. diff --git a/docs/superpowers/specs/2026-05-23-final-parity-execution-spec.md b/docs/superpowers/specs/2026-05-23-final-parity-execution-spec.md index be50dcb..398c257 100644 --- a/docs/superpowers/specs/2026-05-23-final-parity-execution-spec.md +++ b/docs/superpowers/specs/2026-05-23-final-parity-execution-spec.md @@ -91,7 +91,7 @@ Live execution update from this pass: | Area | Current state | Gap to close | Gate | |---|---|---|---| | Core CLI/TUI coding loop | Usable; full tests and 82-case benchmark baseline are green in existing reports | Mostly evidence depth, not missing local primitives | Full test + default benchmark + recent no-stuck dogfood | -| Linux/macOS local CLI gate | TUI entrypoint, task worktree, GitHub fixture smoke, online dogfood, and Linux shell/runtime smoke are available | CI/release matrix now needs to publish macOS shell/runtime smoke results | Non-Windows `agents shell-fixture-smoke --json`, `agents service-smoke --json`, and TUI entrypoint smoke | +| Linux/macOS local CLI gate | TUI entrypoint, task worktree, GitHub fixture smoke, online dogfood, Linux shell/runtime smoke, and PR #14 / CI run #35 hosted macOS shell/runtime smoke are available | Next release matrix run still needs to publish release-binary shell/runtime smoke results | Non-Windows `agents shell-fixture-smoke --json`, `agents service-smoke --json`, and TUI entrypoint smoke | | Model-backed dogfood | Release live gate passed; current live plan reports `105` online runs and `99` successes, with categories `write_validate 29/30`, `recovery 23/25`, `pr_workflow 47/50` | Preserve verified evidence and keep the gate fail-closed in release status | `dogfood report --require-live-runs 100 --require-live-success-rate 90 --require-live-category write_validate:25:90 --require-live-category recovery:25:90 --require-live-category pr_workflow:25:90` | | External write fixtures | `3` disposable real repo online write-fixture samples verified for Rust, Python, and JavaScript; multi-file Python invoice fixture scaffold is now repo-native | Run the new multi-file fixture with an online model and verify evidence | `scripts/create-multifile-external-fixture.sh`, then `dogfood external-fixture ... --evidence-out` plus `dogfood external-evidence --require-successful-external-fixtures 1` | | README real demo | Committed model-backed SVG exists at `docs/demo/deepseek-code-model-demo.svg`, generated from a verified online transcript | Optional polish: TUI/GIF/MP4 capture for launch pages | `record-model-backed-demo.sh`, verifier, rendered media committed | @@ -151,7 +151,11 @@ Live execution update from this pass: - Added to CI/release: non-Windows debug/release binaries now run `agents shell-fixture-smoke --json`, `agents service-smoke --json`, and the multi-file external fixture scaffold smoke. - - Remaining for this milestone: record the hosted macOS CI/release run links. + - Done in PR #14: CI run #35 passed hosted Linux/macOS debug binary + `agents shell-fixture-smoke --json`, `agents service-smoke --json`, and + multi-file external fixture scaffold smoke: + https://github.com/willamhou/DeepSeekCode/actions/runs/26333425574 . + - Remaining for release hardening: record the next release matrix run links. - Windows ConPTY/TCP remains a later cross-platform proof item. 6. Publish and update final public docs. @@ -176,13 +180,14 @@ Do not claim the 5% target while any of these remaining conditions are true: npm/Homebrew publish checks remain credential-skipped, or Windows shell-supervisor ConPTY/TCP evidence has not completed on a real Windows runner. -- For the narrower Linux/macOS local CLI milestone: do not claim closure until - the non-Windows shell/runtime CI/release gates have passed and at least one - online multi-file external fixture evidence artifact is recorded. +- For the narrower Linux/macOS local CLI milestone: hosted non-Windows CI gates + have passed; do not claim closure until at least one online multi-file + external fixture evidence artifact is recorded. Next release-binary smoke + evidence remains release hardening. ## Next Local Action -The next unblocked local action is to run the new multi-file external fixture -with an online model key, verify the evidence artifact, and then record the -Linux/macOS CI/release run links once GitHub Actions has executed the new -non-Windows shell/runtime gates. +The next unblocked local action, once an online model key is available, is to +run the new multi-file external fixture and verify the evidence artifact. The +hosted Linux/macOS CI link is now recorded; the next release matrix run link +should be added after release binaries execute the same gates.