payhumans-payhumans
mcpsmitheryThe human API for AI agents. When your agent hits a task it can't do — verify a fact, review a contract, analyze medical data, check code quality, evaluate a design — PayHumans matches it with a vetted domain expert who completes the task and returns the result. Zero setup required. Call register_agent to get an API key and $50 free credits instantly. No login, no credit card, no approval process. Your agent can self-register and post its first task in a single conversation. Expert domains available: Healthcare & Medical: physician review, clinical data analysis, drug interaction checks, medical literature review Legal & Compliance: contract review, regulatory analysis, IP assessment, legal research, NDA review Finance & Business: financial modeling review, due diligence, market analysis, valuation checks Technology: code review, architecture feedback, security audit, technical writing review, QA testing Research & Academia: literature review, data verification, peer review, fact-checking Marketing & Creative: copy review, UX evaluation, brand analysis, content strategy, design feedback How it works: register_agent — get API key + $50 free credits (no auth needed) search_experts — browse available experts by domain and rate post_job — describe the task, set budget, get matched in minutes get_job — retrieve the completed deliverable complete_job — approve and release payment 9 tools: register_agent · search_experts · post_job · get_job · list_jobs · complete_job · confirm_payment · send_message · get_expert Payments: Stripe (fiat) and Solana USDC. Escrow-protected — payment released only on approval. Integrations: Works with Claude Desktop, Cursor, Windsurf, LangChain, CrewAI, and AutoGen. Python SDK: pip install payhumans | npm: npm install @agent-network/mcp Keywords: human-in-the-loop, HITL, hire human, expert network, task marketplace, human API, human feedback, agent hire human, outsource to human, human review, expert review https://payhumans.co
By Smithery | 34 findings | Scanned 4/19/2026 | tooltrust-scanner/v0.3.8
Risk Summary
Needs ApprovalDep Visibility plus Excessive Permissions raises enough risk that this tool should not be auto-trusted.
Potential impact: This finding indicates the tool should be reviewed before it is trusted.
Recommended action: Keep this tool behind manual approval and avoid unattended runs until the risky capabilities are narrowed or removed.
Suggested policy: keep this tool behind manual approval, do not allow unattended runs, and re-scan after narrowing risky permissions.
Security Findings (34)
tool declares network permission
agents.registerexperts.searchagents.set_webhooksearch_humanstool declares exec permission
jobs.createFix: Tool requests broad permissions (exec/fs/network). Validate input parameters using Enums where possible, and restrict file system operations to explicit allowed directories.
tool declares fs permission
experts.getjobs.createtool declares db permission
search_humansFix: Tool requests broad permissions (exec/fs/network). Validate input parameters using Enums where possible, and restrict file system operations to explicit allowed directories.
tool performs network or execution operations but declares no rate-limit, timeout, or retry configuration
agents.registerexperts.searchjobs.createagents.set_webhooksearch_humansapprove_matchrelease_paymentFix: Declare explicit rate-limit, timeout, and retry configuration for all network and execution tools. Implement exponential back-off and surface resource state to the calling agent.
tool declares http permission
approve_matchrelease_paymentFix: Tool requests broad permissions (exec/fs/network). Validate input parameters using Enums where possible, and restrict file system operations to explicit allowed directories.
Tool did not expose metadata.dependencies or repo_url, so supply-chain coverage is limited.
agents.registerexperts.searchexperts.getjobs.createjobs.getjobs.listjobs.completepayments.confirmmessages.sendmessages.getagents.set_webhookget_task_statussearch_humansapprove_matchrelease_paymentget_pending_actionsraise_disputeFix: Review and remediate the identified issue.
Scan this tool yourself
Reproduce this audit locally, integrate into CI, or let your agent audit its own tools.
Install once, then scan any MCP server:
$ curl -sfL https://raw.githubusercontent.com/AgentSafe-AI/tooltrust-scanner/main/install.sh | bash$ tooltrust-scanner scan --server "npx -y payhumans-payhumans"Adjust the package name if your npm registry name differs from the tool ID. View source
Add badge to your README
Copy this Markdown to show your ToolTrust grade on GitHub.
[](https://github.com/AgentSafe-AI/tooltrust-directory)