A settings card for the DSH LLM auto-retry engine (@deepseek-ai/dsh-llm-retry). Tune the retry count and backoff from Settings → General; changes take effect immediately. Since 0.1.7 it can also auto-continue a reply that was cut off by the output-token limit; since 0.1.9 both the extra error codes and the continuation prompt are yours to write.

中文说明

Features

Includes the settings UI (client bundle lib/client.js): a card in Settings → General — no separate UI package needed.
Overrides maxRetries, initialDelayMs, maxDelayMs, and jitterRatio on the agent/request-error retry policy.
New in 0.1.18 — honors the upstream Retry-After hint: waits what the upstream asks (capped by a configurable limit) instead of giving up the retry; 5 more error codes; logs the hint for diagnosis.
New in 0.1.17 — fixes saves being rejected after a host upgrade (the write is re-read and retried once); the card's build tag now follows the package version; save status sits in a fixed slot so a failure no longer shifts the header; unified fonts in the observation panel.
New in 0.1.16 — the card's build tag is now read from package.json at runtime (it used to be a hardcoded constant that lagged two releases behind); stat tiles are centre-aligned.
New in 0.1.15 — sticky header with save/revert always reachable; the settings card is split into six themed blocks; the observation panel got a clock column, aligned rows and ratio bars.
New in 0.1.14 — declares its host requirement (engines.dsh: >=0.1.6-alpha.1) so the plugin market can show DSH compatibility. Ships every 0.1.13 fix (0.1.13 never reached npm).
New in 0.1.13 — noticeably smoother settings-card scrolling: the scroll container now gets a compositing hint (measured p99 30.2ms → 6.2–12.1ms; frames over 16.7ms dropped from 11–13/410 to 0–2/410).
New in 0.1.13 — model-level overrides no longer silently stop working after a plugin reload: the kernel only re-emits request/header/request/context on the first request and on change (the first-request flag lives on the AgentLoop instance and a plugin reload does not reset it), so sessions that had already made a request resolved no model; the lookup now falls back to the session's own requestContext() / requestHeader(), and the model=(n/a) diagnostics disappear with it.
New in 0.1.13 — settings changes apply immediately: the loader/volatile-update listener now uses { global: true } — the kernel's filter compares a bare fiber against the entry's wrapper, which never matches, so the event was never delivered (the official dsh-llm-deepseek / dsh-llm-pi-ai listeners miss it too).
New in 0.1.13 — no more Chinese leaking into the English UI: all 25 error-code descriptions and the 6 group titles moved into the dictionaries; the settings-page entry name no longer lags behind a language switch.
New in 0.1.13 — smaller fixes: ?tail= normalization on the observation route, jitter-ratio unit, initial-delay > max-delay display, custom error-code input no longer cleared on rejection, a separate continuation prompt when retries are exhausted, and accessibility attributes on chips and inputs.
New in 0.1.12 — auto-continue failed on session format v4: message sources must now be producer-owned, so the continuation carries source.kind = "plugin:dsh-llm-retry" instead of the retired {kind:"plugin"} wrapper; the "continuation landed" line in host.log is detected again.
New in 0.1.11 — compatible with the new kernel 0.1.7-alpha.1 (the settings API moved to SettingsForms: volatile form fields, live config sync via loader/volatile-update, the card reads/writes through remote.settings).
New in 0.1.11 — smoother settings-card scrolling; bug fixes.
New in 0.1.10 — a read-only observation panel at the bottom of the card: retried failures, auto-continues, cap hits and skipped rounds, broken down by error code and provider/model, with the live session models (so you can copy the exact model id into an override) and a log-tail viewer.
New in 0.1.10 — per-provider/model overrides: first matching rule wins, * wildcards, empty numbers inherit the global values.
New in 0.1.10 — backoff curve with a wait budget, a "continue after retries are exhausted" switch (transient errors only), prompt templates, and a zh/en UI that follows the kernel locale.
New in 0.1.9 — add your own error codes straight from the card (the 自定义 group): type a code a provider throws that is not in the known list (e.g. MY_PROVIDER_BUSY) and press 添加. Input is upper-cased and validated (A-Z 0-9 _ - .), duplicates and blanks are rejected with an inline hint, and the code reaches retryableCodes unchanged.
New in 0.1.9 — the auto-continue prompt is editable (continuationPrompt). Leave it empty to use the built-in wording, or write your own line for relays/models that need different phrasing.
New in 0.1.8 — auto-continue now actually fires. Two host-side timing traps: session/event is dispatched inside Session.append, so queueing the follow-up straight from the listener died on session append cannot reenter while another append is being published; once that was fixed, waking the agent at that moment turned out to be silently dropped by the driver (wakeDriver only latches in maintenance/abort), so the message sat in the queue forever. The plugin now waits for agent.whenIdle() and re-checks (new turn started / you sent a message / session disposed) before delivering.
New in 0.1.8 — the host half writes a low-frequency decision log to ~/.dsh/logs/dsh-llm-retry-settings/host.log (capped at 256 KB) and the card points at it. See Troubleshooting.
New in 0.1.7 — error-code list re-audited against the current host build: added PI_AI_NOT_WARMED (adapter warm-up race; a delayed retry usually succeeds) plus three amber "retrying will not help" codes (UNKNOWN_MODEL, UNSUPPORTED_OPTION, REQUEST_EXTENSION). Also clarified TIMEOUT: a stalled SSE stream (stream-idle watchdog) is reported as TIMEOUT — there is no separate code for it, so the host default retry list already covers hangs.
New in 0.1.7 — auto-continue on output truncation (autoContinue, off by default). Hitting the output-token ceiling is not a request failure — the call returns successfully with finish = max-tokens — so no retry policy can ever cover it. When enabled, the plugin watches turn/end and queues one follow-up continuation turn per truncation, at most maxContinuations times in a row (the counter resets when the model finishes normally or you send a new message).
New in 0.1.7 — the error-code chips are grouped into six categories ordered by "will retrying help": transient → rate limit & quota → request & parameters → content & capability → credentials → abort & fallback. Each group shows how many of its codes you selected; picked codes still float to the front within their own group.
New in 0.1.5 — selected error codes float to the front of the chip list, with the unselected ones after them; the order inside each group stays fixed, so chips never jump around when you toggle them. (Since 0.1.7 the floating happens within each category group.)
New in 0.1.3 — configurable retryableCodes: extra failure codes to retry on, merged into each provider's own list (never replaces it). Defaults to INVALID_REQUEST + PI_AI_ERROR, so OpenAI-style HTTP 400 errors (thinking-mode reasoning_text) and generic stream failures get retried out of the box.
Default enabled: false = fully bypassed; nothing changes until you enable the override.

Install

Prerequisite: a DSH Desktop profile (the web profile lives at ~/.dsh/profiles/web).

Option A — GitHub Release package (recommended)

# 1. download the packaged plugin tgz from the v0.1.18 release
gh release download v0.1.18 -R zeng6125-rgb/dsh-llm-retry-settings

# 2. unpack it into the profile's node_modules
mkdir -p ~/.dsh/profiles/web/node_modules
tar -xzf dsh-llm-retry-settings-0.1.18.tgz -C ~/.dsh/profiles/web/node_modules/
mv ~/.dsh/profiles/web/node_modules/package \
   ~/.dsh/profiles/web/node_modules/dsh-llm-retry-settings

# 3. register the bundle in the profile, then restart DSH
#    add "dsh-llm-retry-settings" to dsh.profile.bundles
#    in ~/.dsh/profiles/web/package.json

Option B — dsh CLI / pnpm (requires git + network)

The dsh plugin command forwards its arguments to pnpm in the profile directory:

# from a git repo (pnpm clones it; the committed lib/ means no build needed)
dsh plugin --profile web add github:zeng6125-rgb/dsh-llm-retry-settings

# or from the release tarball URL
dsh plugin --profile web add https://github.com/zeng6125-rgb/dsh-llm-retry-settings/releases/download/v0.1.18/dsh-llm-retry-settings-0.1.18.tgz

Then enable the plugin in the profile: add "dsh-llm-retry-settings" to dsh.profile.bundles (or use the Desktop plugin-inventory UI) and restart DSH.

Note: the command is dsh plugin (a subcommand of the dsh CLI), not dsh-plugin. The dsh CLI is bundled with the Desktop app; if it is not on your PATH, invoke it via the app's node_modules bin, e.g. node "<app>/node_modules/@deepseek-ai/dsh/lib/bin.js" plugin --profile web ....

Option C — from source

git clone https://github.com/zeng6125-rgb/dsh-llm-retry-settings.git
cd dsh-llm-retry-settings
npm install
npm run build        # node scripts/build.mjs → lib/index.js + lib/client.js (no DSH checkout needed)

Link the local build into the profile and register the bundle:

dsh plugin --profile web link "$PWD"
# then add "dsh-llm-retry-settings" to dsh.profile.bundles and restart DSH

Usage

Open DSH Settings → General.
Find the LLM 自动重试 card.
Toggle 开启覆盖 (enabled) to apply the override.
Set maxRetries / initialDelayMs / maxDelayMs / jitterRatio, pick extra retryable codes (chips) — or type a code the list does not know into the 自定义 group — and click 保存.
Optionally toggle 输出截断自动续写 (autoContinue) and set its maxContinuations cap — independent of the retry override above. The line it sends can be rewritten in 续写提示词; leave it empty for the built-in wording.

Changes are written to the dsh-llm-retry settings namespace and picked up live by the retry engine.

Troubleshooting

A truncated reply did not get continued. The host half logs every decision it makes to ~/.dsh/logs/dsh-llm-retry-settings/host.log — one line per event, capped at 256 KB. ctx.logger output is not persisted anywhere in desktop builds, so this file is the only way to see what happened. Read the last lines:

no activate v… line at boot → the plugin is not loaded (check dsh.profile.bundles in the profile).
activate v… names the running build — if it is not the version you installed, the agent was not restarted.
settings registered … autoContinue=false → the switch is off.
a turn/end … line with nothing after it → autoContinue was off at that moment, or maxContinuations is 0.
bail … → the reason is in the line (no agent instance for this session, followup unavailable, …).
放弃投递 … → the agent was still busy when the follow-up was ready, and something changed in the meantime (a new turn started, you typed a message, the session was disposed), so the stale continuation was dropped on purpose.
续写已投递 … chain=N/M → the continuation was queued. 连续续写触顶 means the maxContinuations cap is hit; the chain resets as soon as the model finishes a turn normally or you send a message.

Host-half changes (lib/index.js) need a DSH restart. The client card reloads with the page; the agent does not.

Configuration

Key	Type	Default	Description
`enabled`	boolean	`false`	Enable the retry-policy override.
`maxRetries`	integer (≥ 0)	`2`	Maximum retry count (`0` = do not retry).
`initialDelayMs`	integer (≥ 1)	`500`	Initial backoff before the first retry (ms).
`maxDelayMs`	integer (≥ 1)	`10000`	Upper bound for backoff (ms).
`jitterRatio`	number (0–1)	`0.1`	Random jitter applied to backoff (0 = none).
`retryableCodes`	string[]	`["INVALID_REQUEST", "PI_AI_ERROR"]`	Extra failure codes treated as retryable, merged into each provider's own list.
`autoContinue`	boolean	`false`	When a turn ends truncated by the output-token ceiling, queue one follow-up "continue" turn automatically.
`maxContinuations`	integer (≥ 0)	`2`	Cap on consecutive auto-continuations per truncation (`0` = never continue).
`continuationPrompt`	string	`""`	The line sent to the model after a truncation; empty (or whitespace-only) falls back to the built-in wording.
`continueOnError`	boolean	`false`	Also continue once when a turn ends in a transient error after the retries are exhausted (timeout / transport / server / stream break / empty response). Deterministic failures are never continued.
`overrides`	array	`[]`	Per provider/model policy. Each row: `{ provider, model, maxRetries, initialDelayMs, maxDelayMs, jitterRatio }`; `*` or empty matches anything, and a numeric `-1` inherits the global value. First matching row wins.
`logPath`	string	(host-injected)	Read-only. Absolute path of the host log, injected by the host half for the card's "Open log" button. Not persisted.

How it works

The host half registers the dsh-llm-retry settings namespace (schema validation + persistence + live sync) and prepends an agent/request-error listener that rewrites retryPolicy before the official @deepseek-ai/dsh-llm-retry recover runs:

agent/request-error  →  [this plugin: override count/backoff]  →  dsh-llm-retry recover

The auto-continue half listens to the session log instead, because a truncated reply is a successful request:

adapter finish="max-tokens"  →  agent-loop turn/end{reason:"max-tokens"}  →  [this plugin]  →  agent.followup(continuationPrompt || built-in wording)

turn-stopping cannot be used for this: its payload carries no end reason, so it cannot tell "the model finished" from "the model was cut off". The continuation is sent as a plugin-sourced user message, so the chat renders it as an injected-context row labelled dsh-llm-retry rather than pretending you typed it. Constructor-seeded events (resume / fork / replay) never reach session/event, so reopening an old truncated session does not trigger a continuation.

Which wording works is provider-dependent: some relays do not echo the previous reasoning back on the next request, so the model cannot see where its thinking stopped and may restart the task. If that happens, rewrite continuationPrompt (e.g. ask it to answer directly without re-thinking) — sending more "continue" messages cannot fix it.

UI

Settings → General → LLM 自动重试 card. Edits are draft-based: click 保存 to commit or 放弃 to discard. After saving, the card verifies the write against the settings snapshot and shows 已保存 ✓ / 保存失败 ✗.

The card holds two independent sections: the retry override (count / backoff / jitter plus the grouped error-code chips, with a 自定义 group that accepts free-form codes) and 输出截断自动续写 (its own switch, the maxContinuations cap and the continuationPrompt textarea). A switched-off section is dimmed and its inputs are disabled, but the switch itself stays clickable — turning auto-continue on does not require the retry override to be on.

Requirements

Host plugin: @deepseek-ai/dsh-llm-retry
Client runtime: @deepseek-ai/dsh-client-web-react
DSH settings service

Build

npm run build        # node scripts/build.mjs
npm run typecheck    # tsc --noEmit

License

MIT

Keywords

FAQs

What is dsh-llm-retry-settings?

Is dsh-llm-retry-settings popular?

Is dsh-llm-retry-settings well maintained?

Package last updated on 16 Oct 2026

dsh-llm-retry-settings