MCP server for managing agent-to-merchant micropayment escrow settlements with cryptographic verification and on-chain settlement tracking
This server has severe definition quality issues. Tool 1 (execute_skill_action) has no visible input schema in the source code, only a description. Tool 2 (lock_agent_escrow_settlement) has a schema visible in the spec but the actual tool registration in mcp_server.py does not expose it; the schema appears only in the evaluation metadata, not in the code. Neither tool has descriptions longer than 20 characters that explain WHEN to use them or what they return. No output schemas are documented. The server prints a static JSON response and does not implement the MCP protocol, it merely echoes a status message. No error handling, no parameter validation, no recovery guidance.
Execute micropayment escrow settlement skill action
Lock funds in escrow for agent-to-merchant micropayment settlement with cryptographic verification and on-chain tracking
Tool 1 (execute_skill_action) has NO input schema visible in mcp_server.py.
Tool descriptions are under 20 characters and provide no context for LLM selection. 'Execute micropayment escrow settlement skill action' (54 chars) lacks WHEN to use it, what it returns, or prerequisites.
No output schemas documented. LLMs cannot plan downstream calls or extract required fields. Tool 2 returns a dict with escrow_tx_id, escrow_state, etc., but this structure is not declared in the tool definition.
No error handling or recovery guidance. If a wallet address is invalid or funds are insufficient, the tool provides no actionable error message to guide the LLM's next step.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | F | 24 | 2026-07-28+ | v2 |
Tool names are ambiguous. 'execute_skill_action' is generic and does not convey what action is executed. 'lock_agent_escrow_settlement' is compound (lock + settlement) and could be split into separate concerns.