joekytc/dsh-swarm ↗★ 7

@joekytc/dsh-swarm

A governed swarm of six specialist DSH agents (orchestrator, planner, knowledge-base bridge, developer and two reviewers) that turns a requirement into a strict phase pipeline with machine-verified delivery evidence, review-gated merges, a full audit-log event stream and a live kanban tab; design inspired by the Hermes Agent kanban

패키지
@joekytc/dsh-swarm
호환성
미검증
Harness peer 범위
~0.1.7-rc.2 || ^0.2.0-rc.2
Cordis peer 범위
^4.0.1
버전
0.5.1
라이선스
MIT
최근 업데이트
2026. 10. 9.

설치

$npx -p @deepseek-ai/dsh dsh plugin --profile web add @joekytc/dsh-swarm

Configuration

All keys are optional; schema lives in src/config.ts. Most users only need the first three — keep the rest at their defaults.

The 「模型链」 (model chain) card in the config panel: every role gets its own model — pick provider, model and reasoning effort per role; roles you leave untouched show 「继承」 (inherit) and follow the default.

Model chain card: each of the six roles (v/p/w/d/pt/dt) picks its own provider, model and reasoning effort

KeyDefaultDescription
storageDir$DSH_HOME/storages/kanbanEvent log (events.jsonl), orchestration state, per-task workspaces, dispatcher.log. Value must use the unquoted !!js dshHomePath("storages/kanban") form — quoting degrades it into a literal string
wikiVault.baseUrl'' (empty)wiki-vault HTTP service for KB reads/writes — required for KB features; set to your own server
wikiVault.pagePrefixprojects/Namespace prefix for generated wiki pages
roles.models.{}Per-role model: { provider, model, reasoningEffort?, fallbacks?[] }
roles.models..reasoningEfforthighDefault reasoning effort for all roles
roles.models..fallbacks[]Silent fallback candidates (audited via [model-fallback] comment)
dispatcher.staleTimeoutSeconds14400Heartbeat timeout; running task without heartbeat is reclaimed
dispatcher.maxRetries3Failure retries before circuit → blocked(gave_up)
dispatcher.heartbeatIntervalSeconds300Watchdog heartbeat period
dispatcher.maxProtocolViolations2Protocol-violation guardrail: once consecutive violations reach this many, the next one is final (gave_up)
dispatcher.maxReworksPerRole{ pt: 3, dt: 3 }Max review rework rounds before review/gave-up + [review-final]
prefixRoutes.plan/plan:Command-mode phase-0 planning prefix
prefixRoutes.openspec/openspec:Command-mode approve-and-execute prefix
prefixRoutes.learning/learningLessons / retrospective prefix
prefixRoutes.send/smsFree-delivery prefix (DM with -s)
memory.enabledtrueMemory recall index; false makes planning_memory_recall return a disabled notice
memory.maxIndexEntries8Max recalled memory entries (1–20)
ui.enabledtrueDeclared switch; not consumed yet — the tab registers unconditionally
ui.contentMinWidth715Declared lower width bound (px); not consumed by the client yet — the tab follows the host conversation width
ui.contentMaxWidth780Declared upper width bound (px); not consumed by the client yet
ui.sseHeartbeatSeconds20SSE heartbeat interval
gates.enabledtrueTDD measurement gate; false → silent skip (no event)
gates.timeoutMs600000Per-command gate timeout (ms), SIGKILL at the deadline
gates.forbidden['rm -rf /', 'git push']Command-blacklist substrings (defence in depth)
evidenceReplay.enabledfalseL3 replay of model-written commands — not a sandbox, see Issue evidence check
evidenceReplay.timeoutMs600000Per-replay command timeout (ms)
evidenceReplay.allowPrefixes['npx --no-install vitest', 'npm test', 'npm run build', 'npm run typecheck', 'tsc', 'eslint']Allowlisted tool prefixes (word-boundary match)
imDelivery.enabledfalseWeCom delivery via dsh-im (W3 wrap-up / chain blocked / review gave-up)
imDelivery.botId''Empty = auto-discover the only wecom bot
imDelivery.targetId''Empty = auto-discover the only saved group target
imDelivery.dmTargetId''DM target for /sms -s
imDelivery.fallbackBotId''Bot used when preset-based matching finds nothing
reviewEngine.modedelegatedelegate or managed — see Review engine (ocr)
reviewEngine.managed.provider''Model-chain provider id used in managed mode
reviewEngine.managed.model''Model-chain model id used in managed mode
wikiWritePresets['swarm', 'kanban-w', 'ptc']Presets allowed to call wiki_write (page paths are still whitelisted by namespace)

Configuration notes

  • Mode, provider and model are all chosen on the 「评审引擎(ocr)」 card; the provider/model dropdowns share the same catalog as the model chain;
  • After picking, click 「应用到 ocr」 (Apply to ocr) — the system writes the wiring into ocr's custom config (dsh-managed); the API key is resolved from the dsh model config and written into ocr, never shown in plain text in the panel; if resolution fails it degrades gracefully and points you to a manual ocr config provider in a terminal;
  • When managed is not ready, the ocr tool refuses the managed call and returns delegate guidance; the reviewer switches to delegate mode — nothing is blocked.

Official docs: Installation · Model configuration · Delegate mode