@morlay/session-rdb
RDB durable session backend for DeepSeek Harness: implements both session-persistence and session-branch (rewind / retry / fork) providers.
45 results
RDB durable session backend for DeepSeek Harness: implements both session-persistence and session-branch (rewind / retry / fork) providers.
Provider abstraction + high-level service for rewind / retry / fork over DeepSeek Harness event-sourced sessions.
Rewind / retry / fork orchestration (host) plus conversation.chat.node renderer replacement (browser) — message edit / retry actions on user bubbles.
Settings UI for DeepSeek Harness (DSH) to edit each model provider's request retryPolicy — mode (normal/always), maxRetries, retryableCodes and exponential-backoff parameters — written to settings.yaml and applied by dsh-llm-retry. 模型重试策略设置:在设置面板中可视化编辑各模型提供方的 API 请求失败重试策略。
Execution-discipline plugin suite for DeepSeek Harness: anti-stuck retry interception, forced reflection, todo discipline, memory dedup, and self-evolving skills. Zero dependencies, never touches DSH source.
Whitelist / blacklist / any-error retry policy for DeepSeek Harness, driving DSH's built-in LLM request retry mechanism.
DSH 插件:LLM 自动重试设置卡片——重试次数/退避/抖动与 25 个错误码可勾选(也能自己输入),回答被输出 token 上限截断时自动续写且提示词可改,改完实时生效。
Agent trajectory governance & anomaly diagnosis for DeepSeek Harness: rebuild structured multi-branch trajectory trees from the session/event feed, keep observation-layer snapshots, and run three temporal anomaly strategies (loop deadlock / invalid retry / goal drift).
Agent negative-knowledge ledger for DeepSeek Harness: records failed paths with evidence hashes and retry conditions, auto-invalidates on evidence change.
Weak-network adaptor for the DeepSeek Harness: transparent long-backoff model-stream retries (10-min cap), local response replay cache, heartbeat auto-reconnect, degraded-mode token economy, and a bilingual (zh/en) settings page. / 弱网适配插件:模型流长退避重试、本地响应缓存、心跳自动重连(默认 10 分钟封顶)、降级省 token、双语设置页。
DSH client plugin: 检测回合被中断/出错/超长时, 自动发送「继续」重试, 不做模型/provider 切换。带设置卡片。
Turn-atomic message editing, retry, reroll and version-tree navigation for DeepSeek Harness conversations
Safe model-request retry for DeepSeek Harness with exponential backoff, model-availability recovery, and configurable unknown-error limits.
让 DeepSeek Harness 自动重试商汤 429/QUOTA 等瞬时网关故障直到任务完成——启动时给所有 llm-pi-ai provider 热注入加固版 retryPolicy,运行中新增的 provider 也会自动补上,再也不用手动发「继续」。
Multi-route high availability, fault isolation and observability for DeepSeek Harness. Route groups fail over between authorized provider routes on rate limits (429 / Retry-After), transport errors and auth/quota failures, with per-route quarantine (exponential backoff) and disablement.
Anti-bot HTTP fetch tools for DeepSeek Harness: UA pool rotation, retry with exponential backoff, anti-bot wall detection, and adaptive rate limiting.
失败账本 — DeepSeek Harness 工具失败记录、签名聚类与失败感知重试拦截(含 Web UI 面板)。Failure ledger for DeepSeek Harness: capture, cluster, persist and learn from failed tool calls, with a failure-aware retry guard and a Web UI panel.
Mojibake interceptor for DeepSeek Harness: feature-based garbled-text detection that interrupts garbled writes and asks for review (an identical retry passes).
DSH plugin that auto-retries failed API requests — when third-party APIs return errors (rate limiting, service unavailable, etc.), intercepts and retries to keep the conversation going. Supports auto/manual mode with configurable status codes and error codes.
Auto-continue on 429 rate limit and quota-exhausted errors for DeepSeek Harness Web UI. Host-side RATE_LIMIT/QUOTA recovery plus an on/off switch at the bottom of the chat input area.
高峰自动会话门:周末模式 + 高峰自动暂停 + 官方源二维判定 + 会话级冻结 + 后端自动重试。
LLM 自动重试设置卡片:在 设置 → General 里修改重试次数(maxRetries)与退避时间(initialDelayMs/maxDelayMs/jitterRatio),宿主 dsh-llm-retry 实时生效
DSH plugin: infinite exponential retries on LLM requests, capped at 10 minutes per wait
Fail-closed network recovery, recommended answers, and safe unattended continuation for DeepSeek Harness.