mirror of
https://github.com/deepseek-ai/deepseek-harness
synced 2026-08-15 21:04:50 +00:00
d247c50c6f
The browser e2e lane had been failing wholesale since this stack moved the agent plane into presets, and nothing caught it: 34 of 48 files. Two of the causes are product defects, not test breakage. `bashEnv` goes back to the host plane. `apps/cli/src/web.ts` injects it to publish `DSH_WEB_URL`/`DSH_WEB_MODE`, so the earlier note that "nothing outside the agent plane injects bashEnv" was simply wrong — behind a preset's `shell` realm those variables reached no shell at all, and a `dsh web` agent could not find the address of its own interface. This is the same criterion that returned `subagents`: a host row that injects a service resolves before any session exists and has no agent to key by, so the service is host-plane. `tool-bash` consumes the host registry from inside the preset, which works because an agent context chains to the host; only the reverse is invisible. `tool-subagent-report` goes back with it. It is not a tool this agent calls: it registers a continuable SETUP on the host `subagents` singleton, and that list is not scope-aware. One copy per mounted preset meant every child was handed `report` once per live session, so the second registration threw and a cold subagent resume failed with `subagent-not-resumable` — a diagnostic three layers removed from the cause. The lane's own composition facts follow. Skill roots resolve inside a preset now, a subtree include patches cannot reach, so the scaffold pins the roots' documented environment fallback for its whole lifetime rather than for the boot — presets mount per session. Without it the developer's real `~/.dsh/skills` enters replay requests and goldens while CI sees none. The `apps/cli` composition test pins `storage-json` for the same reason: unpinned it wrote, and then read back, the developer's own `~/.dsh/storages/`. Three tests now address through an agent what they used to read off the root context, because that is where the thing lives: the tool catalog, the skill registry, and the token meter. The seeded-history projection baseline asserts the opposite of what it did — a detached session yields a preset-plane projection only from a durable checkpoint written while it was live, and this seed was written straight to persistence and never ran. Goldens re-recorded for the hero's preset chip and the settings nav entry.
1.6 KiB
1.6 KiB
- dialog "设置":
- navigation:
- text: 设置
- button "通用设置":
- img
- text: 通用设置
- button "模型":
- img
- text: 模型
- button "Agent 预设":
- img
- text: Agent 预设
- button "打开配置文件"
- button "关闭":
- img
- text: 关闭
- heading "模型" [level=2]
- paragraph: 填入各提供方的 API 密钥即可使用其模型。
- list
- text: 提供方
- combobox "提供方":
- option "amazon-bedrock"
- option "ant-ling"
- option "anthropic"
- option "azure-openai-responses"
- option "cerebras"
- option "cloudflare-ai-gateway"
- option "cloudflare-workers-ai"
- option "deepseek"
- option "fireworks"
- option "github-copilot"
- option "google"
- option "google-vertex"
- option "groq"
- option "huggingface"
- option "kimi-coding"
- option "minimax"
- option "minimax-cn" [selected]
- option "mistral"
- option "moonshotai"
- option "moonshotai-cn"
- option "nvidia"
- option "openai"
- option "openai-codex"
- option "opencode"
- option "opencode-go"
- option "openrouter"
- option "qwen-token-plan"
- option "qwen-token-plan-cn"
- option "together"
- option "vercel-ai-gateway"
- option "xai"
- option "xiaomi"
- option "xiaomi-token-plan-ams"
- option "xiaomi-token-plan-cn"
- option "xiaomi-token-plan-sgp"
- option "zai"
- option "zai-coding-cn"
- text: API 密钥
- textbox "API 密钥":
- /placeholder: 输入 API 密钥
- group: 自定义设置
- button "取消"
- button "保存"
- navigation: