Verification and research desk for agents: packet validation, token screens, settlement checks.
Working examples for calling the Ribcage Verification & Research Desk from your agent.
The desk sells work product to agents: claim validation with receipts ($0.25/packet), scoped research (from $2.00), Solana token screens ($0.10 — the deep, cross-source screen), prediction-market settlement verification ($0.05), editing of human-facing text ($5–$15 per 1,000 words), standing watches ($1.00/watch/month), forecast grading ($0.50/log, $5.00 full audit), and settled data by the slice from our own records (from $0.25). Payment in USDC (Base or Ethereum), BTC, ETH, or SOL.
Start free: every caller gets 10 evaluation calls a day with full output. The examples in /examples run against the free tier unchanged.
Sample outputs with the source figures attached are in /samples. Prices live at https://ribcage-agent-services.pages.dev/ and in llms.txt at the same origin. The OpenAPI spec is at https://ribcage-agent-services.pages.dev/openapi.json
Honest scope: settlement verification covers Kalshi and Polymarket markets our data reaches; anything else returns "out of scope," free, rather than a guess. Settled data is drawn from our own work product only — venue data is not resold (Kalshi and Polymarket terms bar it without written authorization).
Independent note: a parallel validation service run by Rowan offers packet validation and research under his own operation. Hire either, or hire both and compare receipts.
Behind every paying agent is a person who funded the wallet and set the spending rule. This desk is built for that person's one question: does the service show its work? Every output ships with the source figures it rests on. Prices are posted and fixed until revised with notice on the service page. A call that fails on our side is not charged. Queued work runs paid jobs first, then evaluation traffic, then internal work.
An MCP wrapper lives in /mcp-server.
Source-derived launch command. Check the maintainer’s required arguments and credentials before running:
npx -y ribcage-agent-services-mcpMerge this template into ~/Library/Application Support/Claude/claude_desktop_config.json. Keep existing servers. Add any arguments, credentials, and permissions required by the maintainer; this template has not been install-tested.
{
"mcpServers": {
"io-github-jrcage-ribcage-agent-services": {
"command": "npx",
"args": [
"-y",
"ribcage-agent-services-mcp"
]
}
}
}Restart Claude Desktop completely for changes to take effect. Confirm the server appears connected in the client’s tool list, then try a read-only example from its documentation.
Claude Desktop setup referenceribcage-agent-services-mcpnpmio.github.JRCage/ribcage-agent-services works with any MCP-compatible client. Copy the config snippet from the Configuration section above and add it to the file shown for your client, then restart the application.
~/Library/Application Support/Claude/claude_desktop_config.jsonRestart Claude Desktop completely for changes to take effect.~/.cursor/mcp.jsonRestart Cursor for changes to take effect..vscode/mcp.jsonReload VS Code window for changes to take effect.~/.codeium/windsurf/mcp_config.jsonRestart Windsurf for changes to take effect..mcp.jsonSave at the project root, then start Claude Code in that project and review the MCP server approval prompt. Keep real credentials out of shared files.