deepseek harness agents

YaoApp/yao

Agent

Delegate tasks to another AI expert via Yao A2A protocol. Use when the user @mentions an expert.

acryldev/acryl

Agent

How the engineering skills should consume this repo's domain documentation when exploring the codebase.

acryldev/acryl

Agent

Issues, specs (PRDs), and Wayfinder maps for this repo live as markdown under specs/. GitHub remote is acryldev/acryl; GitHub Issues are an optional promotion surface, not the default store.

acryldev/acryl

Agent

The skills speak in terms of five canonical triage roles. This file maps those roles to the strings used in this repo's tracker (Status: line on markdown issues, or GitHub labels if a ticket is promoted).

ZSeven-W/dsh-crew

Agent

DSH (DeepSeek Harness) worker on the flash tier - fast and cheap. Delegate simple, well-scoped subtasks: mechanical edits, small scripts, lookups, straightforward fixes.

ZSeven-W/dsh-crew

Agent

DSH (DeepSeek Harness) worker on the pro tier - stronger reasoning. Delegate harder subtasks: multi-file changes, debugging, refactors, anything needing real analysis.

cuddly-guacamole/opencode-routing-suite

Agent

A planning-focused coding agent for designing solutions, reviewing architecture, and planning complex work. It allows reading project information freely but asks before editing or writing files.

cuddly-guacamole/opencode-routing-suite

Agent

You are a helpful software engineer assistant.

shinjiyu/holdem

Agent

A team workflow for splitting an issue into smaller coding tasks. Other agents take tasks, work on separate Git branches, submit merge requests, and the lead agent combines the changes into the main branch.

shinjiyu/holdem

Agent

A description of a software project's roles and hosting model. It says the project has no separate user interface and is operated through hosts such as Cursor, DSH, Codex, or a command-line interface.

shinjiyu/holdem

Agent

Lead: ADL + 拆 Issue（强说明 + 路径互斥） ↓ module/integrator 认领 → 分支 → 实现 → MR ↓ Reviewer → Lead merge.

qinyre/dsh-plugin-capabilities

Agent

Analyze blind comparison results to understand WHY the winner won and generate improvement suggestions.

qinyre/dsh-plugin-capabilities

Agent

Compare two outputs WITHOUT knowing which skill produced them.

qinyre/dsh-plugin-capabilities

Agent

Evaluate expectations against an execution transcript and outputs.

Sivan757/dsh-agent-plugins-market

Agent

Delegate a task to Codex.

Simidas/dsh-compat

Agent

Research the provided topic.