Human-in-the-loop approvals

Engineering operating model for teams where AI agents generate a large share of implementation output.

affaan-m/ECC

Complete feature-implementation workflow for MacMagazine with pattern matching, review gates, and quality verification.

MacMagazine/app-iOS

Rules for working behind Iron Proxy: connect external accounts without handling raw tokens, treat blocked requests as policy outcomes, and never claim a connection before it works.

nanocoai/nanoclaw

Handles a real email account through query and mutate tools: check the inbox, read, search, reply, forward, compose and organize, with sending always confirmed first.

OpenLoaf/OpenLoaf

Build stateful, durable agent workflows with LangGraph. An agent skill from langchain-ai/docs.

langchain-ai/docs

Triages incoming issues and pull requests for a maintainer: verifies claims, reuses past decisions and delivers a docket of open questions, and never merges.

Yeachan-Heo/oh-my-claudecode

Checks whether a defect report has enough context to be accepted, then drafts candidate-only root-cause evidence for human review without confirming causes or changing code.

protect-my-hair/nucleus-marketplace

Draft and approve Reddit comments from scouted opportunities with a hard human-in-the-loop gate.

shawnla90/gtm-coding-agent

Guide for tool registration and tool UI in assistant-ui. An agent skill from compozy/compozy.

compozy/compozy

Opens Plannotator's annotation UI on a markdown file, HTML file, URL or folder, then reads the decision and feedback you send back and acts on it.

backnotprop/plannotator

Removes a personal fact or past event from Muse's memory and from reminders, summaries and scheduled work that could bring it back, after you confirm a plan.

win4r/MuseAI-Skills

Splits substantial coding or repository-review work between a Sol commander that decides and verifies and a separate Luna Max worker that implements, with an auditable run log.

majiayu000/spellbook

Combines the evidence from completed Nucleus workflow steps into a final evidence report and result package, without scheduling steps, updating trackers or opening pull requests.

protect-my-hair/nucleus-marketplace

Shows how to build LangChain agents with create_agent, define tools, add a checkpointer and use middleware for human approval and error handling, in Python and TypeScript.

langchain-ai/langchain-skills

Asks the user mid-run from inside an FSM action or Python type: user agent via the elicitation tools — requesttoolaccess for allowlist approvals and requesthumaninput for HITL (human-in-the-loop)…

friday-platform/friday-studio

Build multi-agent AI systems with LangGraph — the low-level orchestration framework for stateful, graph-based agent workflows.

magnus919/agent-skills

A skill your agent uses when classifying a slice closeout (auto-continue / human gate / park), routing a real decision to a human, or designing a human queue/dashboard surface.

mvschwarz/openrig

Plan and run programs using the SAFe x AI-DLC fusion. An agent skill from bybren-llc/safe-agentic-workflow.

bybren-llc/safe-agentic-workflow

Analyze permission prompts and auto mode denials, then generate permission rules and autoMode entries that cut prompt fatigue.

rohitg00/pro-workflow

INVOKE THIS SKILL when you need human-in-the-loop approval, custom middleware, or structured output.

langchain-ai/langchain-skills

Walk the owner through Mortiflix, the motion design studio on their own machine, from inside Claude Code - install, setup, writing a brief, running the studio, and every review gate through to the…

GTKottman/mortiflix-oss

LLM-assisted human-in-the-loop review. An agent skill from delorenj/mcp-server-trello.

delorenj/mcp-server-trello

A skill your agent uses when addressing Basic Memory pull request feedback, failed checks, or Codex review blockers, then re-running the latest-head review gate.

basicmachines-co/basic-memory

When the user wants to improve their app's onboarding experience, increase activation rate, reduce Day 1 drop-off, or optimize the first-run flow.

appeeky/aso-skills

Build resilient, long-running, multi-step applications with AWS Lambda durable functions with automatic state persistence, retry logic, and orchestration for long-running executions.

awslabs/agent-plugins

Expert in LangGraph - the production-grade framework for building stateful, multi-actor AI applications.

sickn33/agentic-awesome-skills

Human-in-the-loop approval workflows for governed agent actions

ucsandman/DashClaw

Specification generator - 7 phase document chain producing product brief, PRD, architecture, epics, and issues with Codex review gates.

catlog22/Claude-Code-Workflow

Human-in-the-loop approval workflow for high-risk agent operations (file deletions, deployments, config changes).

nwiizo/ccswarm

Prevent destructive operations using Claude Code hooks. An agent skill from rohitg00/pro-workflow.

rohitg00/pro-workflow

Implements defense-in-depth controls at an AI agent's tool-invocation boundary using tool allowlisting, least-privilege identity binding, NeMo Guardrails policy enforcement, human-in-the-loop…

mukul975/Anthropic-Cybersecurity-Skills

A skill your agent uses when you finish an analiz report - the human must approve the analysis before any implementation task is created, via the analizreview column

makifbaysal/tasktrooper

Use BEFORE any spec, plan, or code exists — the Stage 0 gate that decides whether a project should start at all.

LichAmnesia/lich-skills

Specification generator - 7 phase document chain producing product brief, PRD, architecture, epics, and issues.

catlog22/Claude-Code-Workflow

Configure human-in-the-loop gating for AI agent review actions in Claude Code.

wshobson/agents

Execute an approved GRACE 4 GraceChangePlan in sequential or parallel-safe mode with recovery-aware preflight and centralized durable apply.

osovv/grace-marketplace

Human-in-the-loop web studio to tune AI-generated output by eye.

sickn33/agentic-awesome-skills

适用于这类请求：直接在当前 checkout 完成开发、固定做两轮自我 review/refactor、先把最终 review 结果给人类确认、确认后再继续改动、提交或进入下一步。用户可能会说："先开发再自审两轮"、"先 review 两次再给我确认"、"不要开 worktree，直接改"、"先输出 review 结论不要继续"、"做完先停在 gate"。

penwyp/ClaudePreference

Take a working agent prototype to a scalable, observable production deployment on Microsoft Foundry.

microsoft/ai-agents-for-beginners

Take one working agent prototype go scalable, observable production deployment for Microsoft Foundry.

microsoft/ai-agents-for-beginners

The Inkline Guild's PR review checklist — correctness, simplicity, surgical diffs, changesets, doc freshness, semver, and Inkline-specific contract risks.

inkline/inkline

Build and evolve agentic web apps that pair a CopilotKit React frontend with a Microsoft Agent Framework agent running as a Microsoft Foundry hosted agent, connected over the AG-UI protocol.

github/awesome-copilot

Local demo: City-side ADU plan review. An agent skill from mikeOnBreeze/cc-crossbeam.

mikeOnBreeze/cc-crossbeam

Plan and run programs using the SAFe x AI-DLC fusion. An agent skill from bybren-llc/safe-agentic-workflow.

bybren-llc/safe-agentic-workflow

Plan and run programs using the SAFe x AI-DLC fusion. An agent skill from bybren-llc/safe-agentic-workflow.

bybren-llc/safe-agentic-workflow

Establish review gates — criteria, checkpoints, and approval flow.

Owl-Listener/designer-skills

Core scientific methodology principles: research planning, method justification, assumption checking, and human-in-the-loop decision making for cognitive science and neuroscience

NeuroAIHub/BrainPilot

Handle a small fix or maintenance task locally end-to-end — no PR, but full review gates and a confirmed local merge into main

gittower/git-flow-next