MCP Server: io.github.jdhart81/verdigraph

The Verdigrath MCP server provides verifiable cognition by mapping an “agent file” to a deterministic, content-addressed brain_id. Its description states “any agent file” yields a deterministic brain identifier, and the server exposes four tools.

🛠️ Key Features

• Deterministic, content-addressed brain_id derived from an agent file
• Total tool count: 4
• Focus on a five-step carbon and compliance workflow: measure → account → disclose → claim → scan
• Provides multi-agent problem solving and an MCP security preflight
• Paid HTTP routes return structured JSON plus a payment receipt

🚀 Use Cases

• Running the carbon and compliance workflow (measure/account/disclose/claim/scan)
• Multi-agent problem solving
• Performing an MCP security preflight

⚡ Developer Benefits

• Deterministic brain_id for agent files (“verifiable cognition”)
• Structured JSON responses with payment receipts
• Payment gating for paid routes (first call pricing supported)

⚠️ Limitations

• Seven deterministic paid HTTP routes are described, while the tool count is listed as 4
• First paid call from a new wallet is $0.01; later calls use unchanged list prices (list not provided)
• “Safest first paid call” selects only “Regulatory Rada” (details limited)