opl-ledger-evidence-curator

Assess evidence sufficiency and classify gaps for OPL work claims.

8|5|Updated Apr 2, 2026
One-click install
npx skills add https://github.com/gaofeng21cn/one-person-lab --skill opl-ledger-evidence-curator
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: opl-ledger-evidence-curator
Source: https://github.com/gaofeng21cn/one-person-lab/tree/main/plugins/opl-foundation-skills/skills/opl-ledger-evidence-curator
Command: npx skills add https://github.com/gaofeng21cn/one-person-lab --skill opl-ledger-evidence-curator

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

When working on long-running OPL projects like grants, research papers, or patents, you need to assess whether existing evidence supports your claims, identify missing evidence, and prepare evidence briefs, but you are not authorized to write to the append-only ledger or sign owner receipts. This Skill provides a structured, authority-respecting workflow to handle all these evidence curation tasks without overstepping defined boundaries.

Core Features & Use Cases

  • Claim Evidence Assessment: Evaluate claims such as closeout proof, claim support, provenance chains, and receipt review against existing refs to return a clear verdict (supported, partially supported, unsupported, or wrong evidence class).
  • Evidence Classification & Gap Analysis: Organize all existing refs by evidence type (source, execution, owner, runtime, provenance) and identify gaps with clear instructions for what evidence the owner needs to provide next.
  • Use Case: For a research paper closeout, use this Skill to review all commit SHAs, test outputs, and owner receipts to confirm which parts of the closeout proof are supported, and generate a list of missing evidence for the principal investigator to provide.

Quick Start

Use the opl-ledger-evidence-curator skill to assess the evidence sufficiency for the grant closeout proof claim using the existing receipt IDs, commit SHAs, and test output refs from the project workspace.

Frequently Asked Questions about opl-ledger-evidence-curator

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I assess evidence sufficiency for grant closeout claims using existing receipts and commit SHAs?

Evidence sufficiency assessment evaluates existing refs like commit SHAs and receipts to return clear verdicts for grant closeout claims. It categorizes evidence as supported, partially supported, unsupported, or wrong evidence class for compliant brief preparation.

What is evidence gap classification and how does it work for research project handoffs?

Evidence gap classification organizes existing refs by type including source, execution, owner, runtime, and provenance. It identifies missing evidence and generates required owner actions to ensure complete handoff and closeout proof preparation.

Can I prepare an evidence brief without write access to the append-only ledger?

Yes, evidence brief preparation works without ledger write access or owner receipt signing. The workflow assesses claims and classifies gaps using only existing refs, respecting authority boundaries while producing structured claim verdicts and non-authority caveats.

How do I review provenance chains for patent submission claims?

Provenance chain review validates patent submission claims by evaluating existing refs against required evidence types. It produces structured output with claim verdicts, grouped evidence refs, and gap classifications with required owner actions.

What does a wrong evidence class verdict mean when reviewing test outputs and runtime readbacks?

A wrong evidence class verdict indicates the provided test outputs or runtime readbacks do not match the required evidence type for the specific claim. The gap classification output specifies what correct evidence the owner needs to provide next.

What are the limitations of using existing refs for OPL claim support assessment?

The assessment is limited to existing refs like receipts, commit SHAs, and test outputs. It cannot write to the append-only ledger or sign owner receipts, producing non-authority caveats to ensure evidence brief preparation stays within defined boundaries.