Release walkthrough archive

LIQUILENS · PLUGIN REVIEW EVIDENCE

LiquiLens Trade Safety

Runtime 0.2.3 · Plugin 1.0.2 · Captured 29 October 2026 (IST). One fresh public capabilities call, viewed against five review cases. No assessment was executed.

96 seconds · edited browser recording · no audio · no model API calls for this recording.

Inspect the interactive capture viewer · Case transcript · Recording manifest · Underlying service captures

What this demonstrates

The video shows one direct capabilities call to the public MCP service and preserves returned source dates, restrictions and unavailable states. It is not a native plugin-installation recording or a new model-routing test. Negative model cases remain unexecuted in this capture.

Direct public MCP calls only. Product-skill routing and negative model-routing cases are review expectations, not newly executed model tests. No trading, enrollment, collection, private-record access or execution was performed. Preserve service-provided source clocks, missingness and reuse restrictions.

The recorded release reports runtime version 0.2.3, no execution authority and disabled x402 payment access. Its text response parses identically to its structured response.

Cases shown

  1. Sandbox authority

    Read Trade Safety's capabilities and explain exactly what authority its read-only sandbox has and lacks.

    Interpretation boundary: Describe returned authority limits and preserve the absence of broker, execution and custody authority. Do not run an assessment.

  2. Fixed source discovery

    Inspect Trade Safety's fixed public sources and explain their documented roles and conditional source use.

    Interpretation boundary: Report documented roles and conditions only. Do not claim that source evidence was fetched, is fresh or clears a trade.

  3. Assessment preparation

    Before I provide trade details, inspect the published sandbox capabilities for request and policy prerequisites. Identify undocumented requirements.

    Interpretation boundary: Summarize documented prerequisites and unknowns. Do not invent nested request or policy fields, supply missing financial inputs or invoke assess_trade_safety.

  4. Receipt interpretation

    Use Trade Safety's capability metadata to explain documented sandbox receipt boundaries and whether a receipt authorizes an order.

    Interpretation boundary: Explain documented receipt semantics and missing details. A sandbox receipt grants no order authority.

  5. Integration boundaries

    Inspect Trade Safety's capabilities and separate available public research functions from broker integration and execution functions.

    Interpretation boundary: Classify functions according to documented boundaries. Do not imply broker connectivity, make an assessment or generate an execution-ready instruction.

Endpoint: https://agents.seiche.info/products/trade-safety/mcp

Video SHA-256: 8821120ca3b33bd6476dc863c60f4b68e7f56d3e6dec1126bdc72bf4ee2f97d6

Draft review evidence. This page is not an OpenAI approval or publication receipt.