Individual Submission K. Tsoi Internet-Draft Aevum Network Intended status: Informational 19 May 2026 Expires: 20 November 2026 WLP: Wet-Lab Protocol -- AI-to-Automation Instruction Schema v0 draft-aevum-wlp-00 Abstract This document specifies WLP v0, the Wet-Lab Protocol for translating AI hypotheses (PACR records) into machine-executable wet-lab instructions. WLP is the longevity-research analogue of SCP (Science Context Protocol) from the photoresist automation domain. The protocol is royalty-free (Apache-2.0). Reference implementations in Rust (crates/wlp-conformance) and Python (agentcard_adapters.wlp_conformance) are provided. Status of This Memo This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79. Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet- Drafts is at https://datatracker.ietf.org/drafts/current/. Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as "work in progress." This Internet-Draft will expire on 20 November 2026. Copyright Notice Copyright (c) 2026 IETF Trust and the persons identified as the document authors. All rights reserved. This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (https://trustee.ietf.org/ license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Tsoi Expires 20 November 2026 [Page 1] Internet-Draft WLP May 2026 Table of Contents 1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . 2 2. WLP-instruction Schema v0.1 . . . . . . . . . . . . . . . . . 2 3. WLP-outcome Schema . . . . . . . . . . . . . . . . . . . . . 3 4. AgentCard Extension . . . . . . . . . . . . . . . . . . . . . 4 5. Conformance Requirements . . . . . . . . . . . . . . . . . . 4 6. Execution Flow . . . . . . . . . . . . . . . . . . . . . . . 5 7. Lambda Throughput Ratio . . . . . . . . . . . . . . . . . . . 5 8. IANA Considerations . . . . . . . . . . . . . . . . . . . . . 6 9. Security Considerations . . . . . . . . . . . . . . . . . . . 6 Author's Address . . . . . . . . . . . . . . . . . . . . . . . . 6 1. Introduction WLP closes the loop between AI hypothesis generation and physical experimental execution without human translation. Each WLP- instruction carries a PACR provenance link and a lambda (energy) budget, enabling full cost accounting and falsifiability tracing. Key design principles: (1) AI-to-Machine direct translation -- WLP- instructions are machine-executable; (2) PACR traceable -- every instruction has a pacr_record_id provenance link; (3) lambda budget constrained -- platforms MUST reject over-budget instructions; (4) Conformance verifiable -- any platform can run the Conformance Suite. 2. WLP-instruction Schema v0.1 Tsoi Expires 20 November 2026 [Page 2] Internet-Draft WLP May 2026 { "wlp_version": "0.1", "instruction_id": "WLP-{gene}-{yyyymmdd}-{seq}", "target": { "gene": "FOXO3", "itt_l_id": "ITT-L-001", "h_itt_v2": 0.72 }, "intervention": { "type": "crispr_knockdown | compound | overexpression | sirna | small_molecule", "reagent": "sgRNA-FOXO3-001", "dose": {"value": 10, "unit": "nM | uM | mg/kg"}, "window": {"start_day": 0, "end_day": 14} }, "readout": { "primary": "S_T_delta", "secondary": ["lifespan_extension_pct", "healthspan_marker"], "model": "C_elegans | mouse | organoid | cell_line", "timepoint_days": [7, 14, 30] }, "expected": { "S_T_delta": 0.05, "confidence": 0.65, "basis": "superlearner_dqn_v2" }, "constraints": { "lambda_budget_j": 500, "max_duration_days": 30, "min_replicates": 3 }, "provenance": { "generated_by": "II:superlearner-meta-v1", "pacr_record_id": "pacr-discovery-v3:line-XXXXXX", "wlp_conformance": "v0.1" } } 3. WLP-outcome Schema Tsoi Expires 20 November 2026 [Page 3] Internet-Draft WLP May 2026 { "instruction_id": "WLP-FOXO3-20260519-001", "executed_by": "wet-lab-node-001", "execution_date": "2026-05-26", "result": { "S_T_delta_observed": 0.048, "S_T_delta_expected": 0.050, "prediction_accuracy": 0.96, "replicates": 3, "p_value": 0.023 }, "lambda_actual_j": 420, "falsification_triggered": false, "notes": "" } 4. AgentCard Extension Wet-lab nodes MUST declare capability via AgentCard extension fields: { "execution_capability": ["crispr-screen", "organoid-assay", "model-organism", "compound-screen"], "wlp_version": "0.1", "throughput": { "assays_per_week": 10, "lambda_capacity_j_per_week": 5000 } } 5. Conformance Requirements An implementation is conformant if it passes all 5 tests. Reference test vectors are in crates/wlp-conformance/src/lib.rs. Tsoi Expires 20 November 2026 [Page 4] Internet-Draft WLP May 2026 +===================+===========================+ | Test | Acceptance Criterion | +===================+===========================+ | schema_valid | WLP-instruction has all 8 | | | required top-level fields | +-------------------+---------------------------+ | pacr_traceable | provenance.pacr_record_id | | | is non-empty | +-------------------+---------------------------+ | lambda_bounded | lambda_actual_j <= | | | lambda_budget_j | +-------------------+---------------------------+ | outcome_parseable | WLP-outcome has all 5 | | | required fields | +-------------------+---------------------------+ | s_t_delta_range | S_T_delta_observed in | | | [-1.0, 1.0] | +-------------------+---------------------------+ Table 1 6. Execution Flow Superlearner DQN -> generate WLP-instruction -> write to ~/.eon/bus/wlp-queue.jsonl | v wlp_dispatcher.py -> match wet-lab-node AgentCard (execution_capability filter) -> send WLP-instruction to node | v Automated wet-lab platform -> execute experiment -> return WLP-outcome.json | v wlp_outcome_bridge.py -> convert to PACR record (domain="wet_lab") -> write to ~/.eon/pacr-wet-lab-v1.jsonl 7. Lambda Throughput Ratio lambda_throughput_ratio = PACR_virtual_hypotheses / WLP_executed_instructions Target (Day 90): filtered hypotheses (H_ITT_v2 >= 0.5) / WLP count >= 1000x Tsoi Expires 20 November 2026 [Page 5] Internet-Draft WLP May 2026 8. IANA Considerations This document has no IANA actions. 9. Security Considerations WLP-instructions carry lambda budgets that MUST be enforced by execution platforms to prevent resource exhaustion. PACR provenance links MUST be validated before execution. Author's Address Kwai Lap Tsoi Aevum Network Email: kwailapt@gmail.com URI: https://github.com/kwailapt/EON Tsoi Expires 20 November 2026 [Page 6]