Answers hard questions by staging a formal debate: two adversarial subagents argue opposite sides with research, then the main agent rules on the transcript through the debate_open, debate_round, and

Verifiable issue-to-PR delivery chain: an 11-stage pipeline (analysis, search, code understanding, hypothesis, planning, implementation, patch ledger, tests, review, failure classification, PR build)

No description yet

by liujianqiao701

DSH plugin for test engineers: turn an empty folder into a "test work framework" skeleton with one click, and bring framework constraints into the AI session — split requirement docs into modules / confirm versions / cross-document linking / register test environment and account credentials. AI test-framework plugin for DeepSeek Harness.

by Phocinae

Deterministic-first triage for DeepSeek Harness: a rule prefilter classifies every ticket, and anything no rule can place is needs_review rather than guessed at.

by weixshaw

Task Runner Task Decomposition Mode: The main agent decomposes tasks, sub-agents (defaulting to a local model) execute in parallel and then synthesize the results; concurrency and model selection can be configured according to your machine's capabilities. Decompose-and-dispatch agent preset for DeepSeek Harness.

Bot Mode for DeepSeek Harness: a roster of named bots with avatars, personas and per-bot chats, each backed by one durable continuable subagent session that keeps memory across turns.

Deterministic step governance for DSH agents: rejects over-budget agent steps and denies risky tool executions before they run, with shadow mode by default and fail-open on errors.

ICVO coding-standards kit for DeepSeek Harness: a bundle plugin that registers tools on load and injects the standards only when apply_coding_standards is called, plus a P0 gate CLI (npx dsh-coding-ki

by Z-Asset

Data stage — data discovery and quality assessment for asset pricing and machine learning research. Ships one skill.

No description yet

Scheduled tasks and message push manager: cron scheduling with auto push to Feishu and multi-channel, dual-env (OpenClaw/DSH).

by iguowz

Low-Cost Multi-Model Orchestration Plugin (Cortex): Large model handles planning and verification, sub-agents use small models for execution — reducing cost while maintaining quality

by nicecx

Tiered task queue for DSH↔Hermes: queue.json single source of truth, lease/claim model, concurrency-1; Hermes-side cron consumes, DSH-side enqueues. Includes busy-mutex for the approve fast path.

Cordis-native, evidence-driven full-stack engineering workflow: inspect-first planning, explicit verification evidence, and approval-aware gating for sensitive operations.

No description yet

An approval-gated OME engineering delivery preset with session-isolated lifecycle evidence.

Permission mode between workspace-write and full access: shell commands are checked by deterministic rules and a subagent review; irreversible or system-level actions require explicit approval.

A live conversation map for DeepSeek Harness web: real-time session tree with subagent lineage, running state, and one-click navigation.

by zsspub

SQLite-backed personal todo tools and Web panel for DeepSeek Harness

Coordinates PUA and Completion Guard tasks from /task-run or a CLI, requiring frozen acceptance and a current candidate/Goal certificate before completion, with an 8-round and 90-minute ceiling. Verif

by ringoage

The visual, manual subagent-model picker for DeepSeek Harness — predictable per-session routing with global coverage of every in-process subagent path, complementary to auto-routing plugins.

Reusable subagent definitions managed from a Settings page and dispatched through a single delegate_agent tool; ships with a built-in read-only Explorer preset and a nested-delegation policy toggle.

by aa2246740

Durable auto-research experiment loop for DeepSeek Harness, with create / run / monitor in the official Web GUI.