hitl-protocolHITL Protocol — the open standard for human decisions in autonomous agent workflows. When a website or API needs human input, it returns HTTP 202 with a revi...
Install via ClawdBot CLI:
clawdbot install rotorstar/hitl-protocolGrade Fair — based on market validation, documentation quality, package completeness, maintenance status, and authenticity signals.
Sends data to undocumented external endpoint (potential exfiltration)
post → https://api.jobboard.com/searchCalls external URL not in known-safe list
https://github.com/rotorstar/hitl-protocol/blob/main/spec/v0.7/hitl-protocol.mdAI Analysis
The skill describes an open protocol standard for human-in-the-loop decisions and does not contain executable code that sends user data. The external URLs referenced are specification documentation and a legitimate interactive playground, consistent with the skill's educational purpose. No credential harvesting, hidden instructions, or obfuscated behavior is present in the provided definition.
Audited Apr 17, 2026 · audit v1.0
Generated Mar 21, 2026
An AI agent helps a user apply for multiple job postings. When the agent submits applications, the job board API returns HTTP 202 with a HITL review URL. The user reviews the selected jobs, confirms details, and approves the applications through a web interface or inline chat buttons. This prevents applying to inappropriate positions or submitting incorrect information.
A banking AI assistant attempts to process a wire transfer. The bank's payment API returns HTTP 202 requiring human approval. The user receives a secure review link showing transaction details (amount, recipient, purpose) and must explicitly approve or reject the transfer. This prevents fraudulent or erroneous transfers while keeping the user in their messaging app.
An AI travel agent finds flight and hotel options. The booking service returns HTTP 202 with a selection review type. The user reviews the options through a rich UI with cards showing prices, dates, and amenities, then selects their preferred combination. This ensures the user makes informed choices about non-refundable travel bookings.
An insurance AI processes a policy modification request. The insurance system returns HTTP 202 with an escalation review type, routing to a human agent for complex cases. The human reviews the request through a form interface, makes adjustments, and submits the final decision. This handles edge cases where automated processing might be insufficient.
A shopping assistant attempts to complete a high-value purchase. The e-commerce API returns HTTP 202 with a confirmation review type. The user receives inline buttons in their messaging app to approve or cancel the order instantly, or follows a link to review cart details. This prevents accidental large purchases while maintaining convenience.
Offer a hosted HITL Protocol implementation as a service for companies needing human-in-the-loop capabilities. Charge based on API calls, review sessions, or monthly active users. Provide analytics dashboards showing decision times, approval rates, and integration metrics.
Help businesses implement HITL Protocol endpoints in their existing APIs and train their AI agents to handle HTTP 202 responses. Offer custom development for review UIs, security token management, and multi-channel messaging integrations.
Create a directory of services that support HITL Protocol, allowing AI agents to discover human-in-the-loop capabilities. Monetize through featured listings, certification badges, and referral fees when agents use integrated services.
💬 Integration Tip
Start by implementing the polling transport first as it's required, then add optional SSE or webhook support. Use the .well-known/hitl.json discovery endpoint to make your service easily identifiable by HITL-aware agents.
Scored Apr 19, 2026
Meta-skill for AI agent self-improvement. Analyzes runtime logs to detect error patterns, regressions, and inefficiencies, then generates structured improvem...
Stop waiting for prompts. Keep working.
Turn OpenClaw into a learning-loop agent with seeded workspace rules, skill promotion, reflective memory, and proactive maintenance.
Meta-agent skill for orchestrating complex tasks through autonomous sub-agents. Decomposes macro tasks into subtasks, spawns specialized sub-agents with dynamically generated SKILL.md files, coordinates file-based communication, consolidates results, and dissolves agents upon completion. MANDATORY TRIGGERS: orchestrate, multi-agent, decompose task, spawn agents, sub-agents, parallel agents, agent coordination, task breakdown, meta-agent, agent factory, delegate tasks
Complete toolkit for creating autonomous AI agents and managing Discord channels for OpenClaw. Use when setting up multi-agent systems, creating new agents, or managing Discord channel organization.
Billions decentralized identity for agents. Link agents to human identities using Billions ERC-8004 and Attestation Registries. Verify and generate authentic...