Workflow plugins

An opinionated method for modeling work from the user's point of view: user stories with an actor, a need, and a purpose, Gherkin-like acceptance criteria, and routing of technical work, spikes, and enablers. Epic > Story > Sub-task.

1 skill

O Linear da mK como servidor do Claude Code, com o login guardado so na maquina de quem instala. Serve de fonte e de escrita para o mK Produto.

1 MCP server

Behavioral guidelines to reduce common LLM coding mistakes, derived from Andrej Karpathy's observations on LLM coding pitfalls.

1 skill

总工作流开放能力图谱：把 Idea、共创、Goal/Task 开发、PR、产品交付、复盘与叙事连接成可验收闭环。

1 skill

TIA Idea Inbox 与 Global Goal Status Board：收集想法、提升 Goal、同步状态与 Task checklist。

1 skill

双路复盘与三道物理锁：把 Telemetry/Crash 与 Workflow/PR/Infra 证据转成可验收 Task。

1 skill

Run one session as the conductor of the other live Claude sessions in tmux: register them, delegate with the right provider, relay decisions, check RAM before every open, and know when to send the user to the dedicated session instead

1 skill

Think before coding, simplicity first, surgical changes, goal-driven verification.

1 skill

Grice's four conversational maxims as agent guidelines: Quantity, Quality, Relation, Manner

1 skill

Write a self-contained prompt that lets a fresh session pick up the current work cold: goal-driven extraction with verified/unverified claims, dead ends, and a first step. Saves to a collision-free file in /tmp or a repo-configured directory and copies it to the clipboard.

1 skill

Evidence-based AI work supervision: show the executed verification before claiming completion.

PRP workflow, OpenSpec, Vertical Slice Architecture: 37 commands + enforcement hooks.

GitHub PR 생성, Slack 알림, 배포 상태 확인 등 개발 워크플로우를 자동화합니다

Conventions for Angular + Firebase monorepos: shared zod contract, generated Firestore rules, the unit/e2e gate ladder, and the documentation practice.

Harness-native ECC operator layer - 68 agents, 291 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

Writing rules for commit messages, PRs, issues, and docs

Capture what a change is supposed to achieve, keep that goal resident across the session, and verify the work against it before the turn is allowed to end.

Structured interview system for requirements gathering. Conducts focused interviews with templates for feature discovery, API design, database schema, UI components, and technical decisions.

Workflow orchestration for mobile and macOS development - manages TDD cycles, quality gates, visual verification, and context preservation. Enforces hard quality gates including mandatory visual verification with screenshots. Integrates with swiftui-dev for complete development lifecycle management.

Learn Textual basics: app structure, project setup, hot reload, and async patterns

Master Textual app lifecycle: initialization, mount/unmount, configuration, and CLI args

Create reusable, encapsulated custom widgets for Textual applications

Write clear, conventional git commit messages from your staged changes.

F7 self-improvement: a forge-improver subagent that proposes behavior-preserving refinements to ObjectCore's forge scaffolder code, gated by the eval-driven admission pipeline (boundary + golden + guard + full check) before any human merges.

Surface the project knowledge base at session start and prompt capturing durable lessons at session end — the KB's automated read/write surface.

Workflow/saga orchestration, rules, form builder, wizard

Adversarial review of plans and artifacts with personal lead review by default and bounded opposite-provider review on explicit request. A designated writer applies authorized fixes; report-only targets stay unchanged.

Session posture for expensive orchestrator models (Fable/Opus): /advisor puts the main agent into a delegation-only mode - it writes specs, dispatches cheaper implementer agents, personally reviews spec compliance and quality by default (cross-provider review is opt-in), triages findings, and merges, while a hard PreToolUse gate (exit 2, bypass-surviving) blocks its own file edits and non-orchestration Bash. Subagent calls pass untouched; a per-turn posture injection keeps the model delegating instead of fighting denials. ADVISOR_DIRECT=1 one-off hatch. State is per session: every session starts in the mode, and CCGM_ADVISOR_AUTO=false opts out.

[DEPRECATED] Go-based terminal UI for managing Claude Code agent processes via tmux panes. Unmaintained and no longer offered for new installs; a GUI-based replacement is being pursued instead. Kept in-repo for existing users.

Live SIA dashboard for Claude Code: agents at work, tokens handled by subagents and the estimated saving from cheaper models, project skills and how often they are used, and every SIA call. Needs Claude Code 2.1.287 or later.