io.github.meltingpixelsai/zero-core-verify

Measured conclusion: Deep-tested. L2 outcome: tools_ok.

Post-tx outcome verification via LLM-as-judge. Zero Core Verify.

Add to your agent

Copy this into your MCP client config:

{
  "mcpServers": {
    "zero-core-verify": {
      "url": "https://verify.rugslayer.com/mcp",
      "transport": "streamable-http"
    }
  }
}

Measured tools (5)

• list_tools

  List all available Harvey Verify tools with pricing and input requirements. Use this for discovery.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• health

  Check Harvey Verify server status, uptime, and payment network configuration.

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• verify_outcome

  Post-transaction verification. Code checks what it can: the response against expected_schema (a JSON Schema or a list of required field names), and a measured response time against a stated limit. A response that fails code's checks fails without the AI judge; otherwise the judge scores accuracy and completeness as advice. Without a schema or a time limit only the judge decides, and decided_by says so. Returns a verdict (pass, fail or inconclusive), decided_by, scores, format compliance, SLA result, issues, and what code checked.

  Parameters:

  • request_description (required) — string — What was requested from the service
  • response_data (required) — string — The actual response/output received from the service
  • service_id — string — Service identifier for quality tracking (e.g. 'harvey-tools/scrape_url')
  • expected_schema — string — A JSON Schema, or two or more required top-level field names (letters, digits, _ and -) separated by commas, e.g. 'name, price, url'. Anything else is passed to the AI judge only.
  • sla_requirements — string — SLA wording for the AI judge (e.g. 'must include all fields'). For a time limit code can check, use the two fields below.
  • sla_max_response_time_ms — integer — The response time limit, in milliseconds
  • observed_response_time_ms — integer — The response time you measured, in milliseconds

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• get_service_quality

  Get aggregated quality scores for a service based on past verifications and reports, counting each paying wallet once (its latest). Returns the pass rate (overall, and over the rows where code checked the format or a measured time), average completeness and accuracy over the scored rows with their count, the count of inconclusive rows, format and SLA compliance rates, and a trend by pass rate. The scores are self-reported by paying wallets and not attested: nothing proves a reporter used the service.

  Parameters:

  • service_id (required) — string — Service identifier to query (e.g. 'harvey-tools/scrape_url')

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

• report_outcome

  Record a simple pass/fail outcome report for a service call. No LLM analysis - just logs the result to the quality database. Cheaper alternative to verify_outcome when you only need to record success/failure.

  Parameters:

  • service_id (required) — string — Service identifier (e.g. 'harvey-tools/scrape_url')
  • was_successful (required) — boolean — Whether the service call succeeded
  • notes — string — Optional notes about the outcome
  • response_time_ms — integer — Response time in milliseconds

  Hints: read-only not declared · destructive not declared · idempotent not declared · open-world not declared

Publisher website Repository

Measured by MCPYP probing. Untested fields are omitted, never estimated. Listing is always free; MCPYP never proxies calls or stores user tokens.