dsh-vision-autoswitch
DeepSeek Harness plugin: auto-route image-bearing requests to deepseek-v4-flash-vision-exp, then fall back to the original model.
180 results
DeepSeek Harness plugin: auto-route image-bearing requests to deepseek-v4-flash-vision-exp, then fall back to the original model.
DSH task chime: play local sounds for approval/permission requests and task completion, fully configurable from the Web GUI settings. DSH 任务提示音:审批/权限请求与任务完成提示音,可在 GUI 设置中自定义声音、音量与冷却时间。
Runtime-switchable outbound proxy for the DeepSeek Harness: hot-reload HTTP(S)/SOCKS5 routing for every global-fetch request via a settings section.
DeepSeek Harness plugin: pick configured model providers, and when a model fails with a model-related error the request automatically falls through the provider group's models (then the next selected provider) until one works, so the unfinished task keeps going. Adds its own Settings tab.
GitHub issue and pull request references for the DeepSeek Harness web GUI: click the GitHub icon at the bottom-right of the input box to search the workspace repository and insert issue/PR references. Data comes from the gh CLI.
Global model command palette and provider/model configuration panel for DeepSeek Harness, with selective request retries, OpenRouter free-model import, loopback relays, online capacity presets, and optional media tools.
DSH native tool: analyze the current branch diff, generate a Conventional Commits PR title, motivation/scheme/testing/risk description and a self-review checklist, write PR_DESCRIPTION.md, and optionally open the PR via the gh CLI.
Minimal-anchored agent preset for DeepSeek Harness: first request keeps the real Minimal tool pair, promotion unlocks the full Standard catalog with a one-shot background-jobs notice
Offline DeepSeek request, tool-loop, reasoning_content, and SSE diagnostics for DSH
DeepSeek Harness plugin: model sampling sliders (temperature / maxTokens) applied to every provider via the agent/request hook, with live settings persistence.
HTTP/API debugging toolset for DeepSeek Harness: http_request, curl_parse, request_history
LLM pre-review for sandbox-escalation approvals: an independent-context LLM gate answers sandbox escalation requests before they reach the user, falling back to the user on any failure.
Native DSH http_request tool with configurable credential whitelist and usage skill.
Agent loop rate limiter — prevents TPM/RPM limit violations by intercepting the LLM streaming pipeline and adding adaptive delays between requests
X-ray for your DeepSeek Harness — diagnostics for what's actually loaded, why, and what it costs: per-plugin context-tax attribution, per-request token ledger, skill catalog pricing, dependency cascades.
Read-only model request concurrency monitoring for DeepSeek Harness
DSH web plugin: cross-session token usage, request count, and optional cost analytics (ctx.tokenUsageStats), with a self-contained dashboard page and a sidebar footer entry.
Provider/model-aware adaptive pacing and visible rate-limit retries for DeepSeek Harness free models.
Maestro Review — pluggable ReviewProvider (GitLab/GitHub) + orchestrator for automated MR review
Issue-to-PR 可验证交付链:项目配置 → 11 阶段流水线 → 人工复核 → PR 说明(DSH 全局插件)
Keeps Charm Hyper requests under the gateway's 10 MiB body cap: replaces the oldest images with placeholders (newest survive) before sending, and retries with a tighter projection when the gateway still rejects the body.
DeepSeek Harness plugin: automatically adds a stable per-conversation x-opencode-session header to OpenCode / OpenCode Go provider requests — fixes 400 MissingSessionID and keeps OpenCode's session affinity / prompt-cache routing working.
DeepSeek Harness plugin: reports what changed in your authored pull requests since your last check — merged, closed, or gone stale.
Mobile UI fixes for the DeepSeek Harness web profile: a restart row, input-modality switches, a narrow-screen drawer and settings tabs, a full-width model menu, a stuck-request network chip, a scrollable question region, a throttled reaction to streaming message storms, a guarded subagent-catalog refresh, click-to-open lineage chips, a narrow-header collector sheet, and a throttled session-list render under background streaming load.