External AgentScope runner wanted for one bounded agent-message decode

[jaden3824]

I am seeking one AgentScope operator outside the Urusilla project to test a single declarative inter-agent message. This is a bounded interoperability request, not a promotion claim or evidence of adoption.

The current broad result is unfavorable: demonstrated post-decode API-input saving for general unfamiliar-agent dialogue is 0%. Safe fallback to concise text is the main verified general-use behavior so far.

Reproduction surface

• 750-byte packet, SHA-256, expected typed JSON, and decoder command: https://github.com/jaden3824/urusilla/blob/51ff0e3/interop_lab/evidence/challenge_001.md
• One-record Hugging Face copy and pinned result thread: https://huggingface.co/datasets/jaden3824/urusilla-interop-lab/discussions/1
• Structured result issue: Good first evidence: decode challenge_001 in a fresh external runtime jaden3824/urusilla#7

A useful AgentScope result may be an exact decode, mismatch, refusal, fallback, task failure, or null token result. Please report the AgentScope/model versions, exact supplied context, prior exposure, whether the project decoder or model-only reading was used, any repair turns, and all provider-exposed token categories. Unknown categories stay null.

The packet is unsigned, read-only, and non-effect-authorizing. It grants no tools, persistence, cross-session memory, spending, permission expansion, network action, or external effect. No Urusilla package installation is required for a model-only reading attempt. Publication is the operators separately authorized action.

A single run would not prove independent implementation, organic propagation, security, general adoption, or state-of-the-art performance. Negative evidence will remain visible.

Disclosure: Codex agents assisted with the fixture and this post; I reviewed and submitted them.