dsh-request-manager
请求管理插件(bundle):按全局配置管理 Agent→LLM 请求——「请求额外间隔」(每个请求发送前固定多等 N 毫秒,0=不介入)、重试次数覆盖(留空=沿用 provider 默认)、拦截「请求失败」(终态失败后挂起等待人工决议:重试原地重发同一请求 / 放弃按原生路径终结)。入口为会话头部 utilities 槽位的「请求管理」按钮;设置跨会话全局生效,按值决定是否介入;界面跟随 dsh 语言中英切换。
187 results
请求管理插件(bundle):按全局配置管理 Agent→LLM 请求——「请求额外间隔」(每个请求发送前固定多等 N 毫秒,0=不介入)、重试次数覆盖(留空=沿用 provider 默认)、拦截「请求失败」(终态失败后挂起等待人工决议:重试原地重发同一请求 / 放弃按原生路径终结)。入口为会话头部 utilities 槽位的「请求管理」按钮;设置跨会话全局生效,按值决定是否介入;界面跟随 dsh 语言中英切换。
DeepSeek Harness plugin: route an agent request to a vision model (deepseek-v4-flash-vision-exp) when the user message contains an image, and keep the original model otherwise.
DSH plugin: route every fetch/http/https request of the DSH process (model endpoints, web search, MCP) through an HTTP proxy. Follows *_proxy environment variables or manual settings; configured from the Web UI (设置 → 网络代理).
DSH plugin: anonymize workspace-external file paths before model requests with user confirmation
Explicit minimal-solution coding guidance for DeepSeek Harness. Registers six load-on-request Ponytail skills and a localized Settings page.
面向 agent 的 GitHub 工作台:仓库、issue、PR 与代码审查的读写工具集,以 MCP stdio server 形式运行,零运行时依赖,可被 dsh 等插件化 harness 加载
GitHub issue/PR search and view tools (gh_search / gh_view) for DeepSeek Harness, backed by the GitHub CLI (gh). Personal plugin ported from oh-my-pi's gh-search / gh-view — not official.
Session-scoped LLM request and response inspector for DeepSeek Harness Web
DeepSeek Harness plugin: injects an [IMPORTANT] chain-of-thought rule block at the very top of the system prompt of every model request (English CoT starting with I'm, no 'Let me', direct verbs).
SillyTavern feature port for DeepSeek Harness: character cards, world info (lorebook), and generation presets as prompt sections, runtime context, request overrides, and slash commands
Request image optimization, intelligent retention, and observability for DeepSeek Harness (DSH)
GitHub REST API tools for DeepSeek Harness: repositories, search, issues, pull requests, comments, account binding, and workspace integration, with a browser-side GitHub panel in the web GUI
DSH rc.8 Web settings row for active providers' model-request retry budget
Long-term memory for DeepSeek Harness backed by Hypatia: a host-side CLI adapter, a plugin-owned SQLite control ledger, same-request recall, and narrow memory tools. Requires the `hypatia` CLI on PATH.
LiteLLM /v1/search-backed web search provider for the DeepSeek Harness web capability seam (ctx.web). One search is one HTTP request — no model turn.
Configurable auto-approval for DeepSeek Harness: intercepts approval/request and answers allowed-once/rejected by policy (allow-all / allowlist / deny-always), with a full audit log — so the human is only asked when the policy says so.
A DeepSeek Harness Web plugin for inspecting redacted LLM HTTP request and response traces.
Processes tab for DeepSeek Harness: a live per-process console (main agent and every subagent) with raw tool requests, responses and timing, installed as one profile bundle.
Native desktop notifications for DeepSeek Harness (dsh): run completion, model questions, and approval requests on macOS, Linux, and Windows. Includes a web settings card consistent with the dsh web UI.
Bounded all-error request retry plugin for DeepSeek Harness
Local, zero-network, fail-closed secret preflight for DeepSeek Harness model requests.
Tolerate redundant sandbox_permissions requests: a tool call asking to escalate to a mode that is not strictly wider than the session's standing mode runs as-is at the standing mode instead of failing with "not strictly wider". Genuine escalations still route through the approval flow.
Provider request metadata minimization extension for DeepSeek Harness
HTTP 请求工具:发起请求返回状态码/响应头/响应体/耗时(Node fetch)