Workflow plugins

5-phase workflow framework — structured project management with TDD, verification gates, team orchestration, and systematic debugging

13 skills · 12 agents

Project bootstrapping, verification against the real codebase before changing it, evidence-based pre-merge review, and stacked PRs with gh-stack.

23 skills

Spec-driven development chain: constitution, spec, plan, tasks, consistency check, execution, drift detection and amendments.

23 skills

Deterministic quality-gate + dangerous-command guard hooks, a read-only default-FAIL evaluator agent, a Definition of Done, a self-tooling (no-silent-stub) contract, the `cmpl` CLI, idempotent md→Beads sync, and /completely:* slash commands.

8 skills · 1 agent · hooks

Delegate serious multi-file work to Codex CLI under two hard gates: Claude approves the plan Codex authored, Codex implements and accepts follow-ups from the SAME persistent thread, Claude reviews the diff against it. One writer lock sequences implementation, revisions, and prompts; schema-checked verdicts survive codex-cli updates.

5 skills · hooks

Fighter builds, cleric reviews and fixes, wizard advises — plus Session Zero and an experience system that levels up your project's memory. Install once; it is on in every project it is installed for.

2 skills · 3 agents · hooks

Specification-Driven Development: governance, workflow commands, and enforcement hooks for Claude Code projects

1 skill · 3 commands · hooks

Premium Claude Skills: Think Before Coding, Simplicity First, Surgical Changes, Goal-Driven Execution

4 skills

The Oathe claim loop: board at SessionStart, statements as you work, lease heartbeat at Stop, frame note before compaction.

1 skill · 1 command · hooks

Poimandres' Claude Code plugin.

1 skill · 1 MCP server

Audit error handling for silent failures, inadequate feedback, and inappropriate fallbacks using one canonical methodology and the strongest isolated reviewer available.

1 skill · 1 agent

Maintain shared subagent instructions and generate native coding-tool configurations.

1 skill

让主 agent 当中枢调度者：开工前对齐范围与验收口径，之后自主派发隔离子代理、用文档交接、独立验收门把关、节奏化提交到可交付。领域无关，Claude Code / Codex 通用。

1 skill

Surface what I think we're doing before acting on it. Produces a structured manifest exposing every silent commitment so the user can correct course before any artifact is produced.

1 skill

Surface and resolve underspecified decisions before designing. Asks targeted questions about scope, edge behavior, trade-offs, integration, constraints, and audience (with defaults), so design choices aren't silently locked around hidden assumptions.

1 skill

Harness-native ECC operator layer - 68 agents, 286 skills, 94 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses

200 skills · 94 commands · 68 agents · 1 MCP server · hooks

cc-code marketplace：给 AI 立规矩的开发工作流，规划产出断言，dev 一次完整实现；任务清单 tasks.md 分清阶段 / 泳道 / 任务，调度由脚本算、泳道并发用 Workflow；交付前 check-fix 只比对本次改动、当场修，符合才许上线；多人协作按卡隔离（任务分支 + 团队名单）。双端支持 Claude Code / Codex。

19 skills · 6 agents · hooks

Branch-based game Korean-localization workflow for Codex and Claude Code: no-API built-in imagegen best-of-three selection, original-feel Korean image QA, platform-scoped smoke testing, official-distribution direct runtime QA, recoverable cleanup, and release. Complete NSW and Steam/PC adapters plus SFC/PS1/PS2 skeletons and Unity/VN/LucaSystem engine modules.

15 skills

The BBQ flow (Blueprint → Build → QA) — a planning pipeline with blind QA and a visual review surface, plus Linear issue and project drafting, adversarial, intent, and pattern review, and a CLAUDE.md refactor utility.

10 skills · 3 agents

Project development workflow plugin — dynamic soft orchestration via Agent Graph (requirements-incomplete / mid-change / continuous-iteration friendly)

8 skills · 5 commands

Complete modernization workflow with multi-agent orchestration (assess viability → create plan → execute). Includes 6 specialized agents and comprehensive protocol documentation for systematic software development

1 skill · 6 commands · 6 agents

Splits a task into subtasks, hands each one to a predefined agent type and picks Haiku, Sonnet, Opus or Fable for it by the kind of work, then runs the subtasks in dependency waves and verifies once. Skills dispatch and new-agent; agents scout, mechanic, researcher, builder, writer, reviewer, verifier, architect, debugger and specialist; seven task templates, a plan checker that catches overlapping file ownership before any agent starts, a run log whose stats suggest route changes, a plan-to-Workflow compiler and a scaffolder for new agents and agent plugins. Needs Python 3.12+. Lives in its own repo.

2 skills · 10 agents · hooks

sonmat — External verification system. Six verification axes (guard/inspect/witness/punch/devil/scribe) + autoloop orchestration + requery retrieval cycle. Discovery-led depth, protocol-isolated intent/baseline-artifact comparator (witness), post-work persistence layer (scribe).

7 skills · 3 agents · hooks

Workflow skills: write-clearly (plain-language and unambiguous technical prose), replan (iterative planning), review, review-staged, and second-opinion (proof-based code review), gitplan and stage (disciplined Conventional-Commit staging), and the Technical Project Plan workflow (tpp, handoff, tpp-orchestrate — multi-session plan files with serial orchestration, TDD subagents, and cross-model vetted reviews).

11 skills

让 AI Agent 学会使用 data-flow 框架的插件：节点/流程编写、DAG 模式、调试排查、CLI 命令参考。

5 skills

Open source contribution workflow -- upstream sync, CONTRIBUTING.md compliance, code style matching, PR preparation, CLA/DCO signing, AI disclosure, and PR quality bar. Agent: pr-analyzer. Hook: upstream-sync-check. Commands use dynamic context injection.

1 skill · 2 commands · 1 agent · hooks

Clone or reference this repo as a Claude plugin. /harness-init grounds itself by reading repo files; the IDE_Agent reads, judges, and writes root-truth files (AGENTS.md, CLAUDE.md, steering/, Cursor adapters) after one user confirmation. No Node/npm/npx prerequisite for users.

5 skills

Context window management -- compaction timing, context rot prevention, branching strategies, and sub-agent isolation patterns. Agent: context-advisor.

1 skill · 1 command · 1 agent

IW4X/MW2 modding toolkit — IWD archive tools, GSC lint/format/orphan analysis, and a DVAR knowledge base, as MCP tools and the iw4x-toolkit CLI

2 skills · 1 MCP server

Deterministic, resumable JavaScript workflows across Codex, Claude, Cursor, and Kimi agents, with journaled resume and worktree isolation.

1 skill · 1 MCP server