642 tagged human-in-the-loop, measured the same way as everything else here.
Browse within: brainstorming 117context-engineering 117antigravity-skills 116harness-engineering 116local-first 110help-wanted 100openai 100openrouter 100agent 76agent-control-room 76agent-society 76generative-agents 76engineering 75prd 75
arcayne/agentic-delivery-playbook
Skill Claude CodeCodex
Coding Agent Preflight for non-trivial coding work in Pi: classify direct/lightweight/full, lock goal/non-goals/risks/evidence, require approval when needed, and coordinate Pi goals/subagents with truthful routing. Use when ambiguity, risk, scope, or reviewability justifies a delivery contract; skip clear small direct…
dinosn/security-knowledge-base
Skill Codex
Read, search, and inspect evidence in a Security Knowledge Base v1 repository, then validate or stage evidence-cited, revision-bound claim and finding proposals. Use when a task refers to a repository containing kb.json and the kb CLI, asks to check or update the security KB, or supplies an skb.context-packet/v1. Stop…
Skill Claude CodeCodex
A procedure for consulting a frontier-class external reasoning model (extended-thinking tier) on a decision that matters — and for accepting its answer afterwards. Generates the consultation prompt, then audits the reply. Vendor-neutral: any top-tier model will do. Not for routine work sent to a cheap model — that is…
Skill Claude CodeCodex
Human-in-the-loop skill for AI agents. Use when your agent needs humans to review, approve, or complete a task. Provides REST API and MCP (Model Context Protocol) integration with long-polling and webhooks. No server required — agents self-register and get responses back asynchronously.
Skill Claude CodeCodex
Drive the Resonate server from the shell — start a server (serve / dev), create/resolve/search Durable Promises, create and delete cron schedules, manually invoke a function via invoke, walk a call-graph with tree, and inspect or recover tasks (tasks). Reach for this when the agent needs to poke a running Resonate…
chengxi271-commits/autodl-research-pilot
Skill Codex
Inspect, select, rent, and observe AutoDL compute for a concrete research workload. Use when Codex must decide between local, existing, Pro, or Enterprise Elastic GPUs; validate disk, network, persistence, and runtime fit; or operate an approved AutoDL run through the API and SSH.
Skill Claude CodeCodex
A design workbench for turning a vague task into a controlled, repeatable Agent Loop. An Agent Loop is a task process with defined steps, checks, stopping points, and recovery actions.
WendyAi2005/codex-comsol-bridge
Skill Claude CodeCodex
Use this skill for Codex-to-COMSOL automation through MathWorks MATLAB MCP Server and COMSOL LiveLink for MATLAB. It connects two mature, officially supported integration paths into one community-built workflow with native COMSOL model access, task-level local execution, compact result return, same-server Desktop…
Skill Codex
Use Lantern's application MCP to create, draw, compose, revise, organize, inspect, or review a creator's comics through the domain capabilities Lantern currently exposes. Use for comic work in Lantern; do not use for developing the Lantern source repository.
Skill Claude CodeCodex
Use the API harness to drive Agentweaver through its REST API, capture the complete request/response evidence, and emit a normalized agentweaver.persona-judge-verdict/v1 JSON verdict. It is for backend/API end-to-end validation; use the UI or MCP harness for those surfaces.
LiveTrack-X/spec-driven-ai-development
Skill Codex
Install or upgrade SDAD; migrate an existing SDAD project; recover or repair sdad-state.yaml, INDEX, ledger, or handoff consistency; run or interpret SDAD Doctor; or diagnose the SDAD control plane. Use only for SDAD-specific bootstrap, migration, repair, and Doctor operations.
Skill Claude CodeCodex
Operate Atlas when orchestration adds value. Use its eight MCP tools to validate, save, preview, run (sync or async), list runs, inspect, cooperatively cancel, and safely resume interrupted workflows. Always validate and dry-run before a billable run.
Skill Claude CodeCodex
Prepare a Proton Mail draft, optionally with user-supplied attachments, after explicit in-conversation confirmation; the user always reviews and sends it in Proton Mail.
Skill Claude CodeCodex
Use when Codex needs read-only external feedback for planning, architecture review, code discussion, strategy review, or decision support.
Skill Claude Code
Render a proposed UI change or set of design concepts as real HTML in the project's own design system, serve it locally, and loop on the operator's clicked decisions and inline annotations until they approve. The visual alternative to describing UI in a wall of terminal text or a markdown plan. Use when the operator…
Skill Claude CodeCodex
Respond to a human's anchored Tether comments on a markdown draft by PROPOSING rewrites (suggestion mode) — the human then Accepts or Rejects each in the editor. Use when the user says "apply my Tether comments", "address my comments", "suggest fixes for my comments", or points you at a .md they've commented on with…
baixinghao/intent-gate-service
Skill Claude CodeCodex
A procedure for sending an unresolved decision to a DingTalk group, where people can answer through the bot. DingTalk is a team messaging platform commonly used in China.
Skill Claude CodeCodex
Hire a verified human through Human For AI (humanforai.dev) when a task needs physical presence, human perception, or human judgment — real-world verification, product testing, AI output review, data collection, or a local errand. Free during the pilot, no auth.
Skill Claude CodeCodex
Run a complete Handover MCP continuity test across separate authenticated people or service agents. Use when asked to test an agent handoff end to end, qualify a Handover integration, verify multi-file round-trip, exercise revision-anchored review and correction, or test denied, read-only, stale-revision, and…
hsuan1012/Dataviz-Coding-Agent
Skill Claude CodeCodex
A workflow for building interactive data-visualization apps from CSV, JSON, or GeoJSON data using React, Tailwind, Chart.js, or D3.
Skill Claude CodeCodex
Safely retrieve readable text from one public URL for SimonOS. Use when a user sends a single http(s) link and asks to read, summarize, or inspect it; return a clear limitation for private or unsupported social sources.
Skill Claude CodeCodex
How to compute what an access grant actually unlocks, rather than what its policy document says. Load before writing or reviewing any ACCESSGRANT change.
Skill Claude CodeCodex
A price-monitoring workflow for Amazon.sa links that includes human approval and Telegram alerts.
Skill Claude CodeCodex
Talk to your human over Beacon while you work. Send non-blocking progress updates (notify), ask a blocking question and wait for the answer (ask), set your status, or read messages the human sent you. Use when running a long task and you want to keep a human in the loop, or need a decision before an…
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: