agentsclimarketplace

Settlement witness skill

Skill nutstrut/settlement-witness-skill

Verify structured agent outputs and return deterministic verdicts with signed receipts and optional reputation attribution.From its SKILL.md

Install
npx -y skills add nutstrut/settlement-witness-skill

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

2 things to look at

  • no licenseNo license file was found in the repository. Code published without one is not open source by default, so using it at work is a question for whoever answers licensing questions where you are.
  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

SKILL.md

2.0 KB, 438 tokens by cl100k_base, as published. Nobody here has run it

SettlementWitness

Verify structured agent task outputs and return signed proof of outcome.


Core Principle

Verification should be deterministic, minimal, and optional.

Only structured inputs are evaluated.


What This Does

SettlementWitness evaluates submitted task data and returns:

  • PASS / FAIL / INDETERMINATE
  • signed receipt (SAR)
  • optional TrustScore update

Core Execution Loop

  1. Provide input:

    • task_id
    • optional agent_id
    • spec (expected result)
    • output (actual result)
  2. Submit for verification

  3. Receive:

    • verdict
    • receipt
    • optional score update

Example

settlement_witness({ task_id: "task-001", agent_id: "0xYourWallet:agent", spec: { expected: "report generated" }, output: { expected: "report generated" } })


Result Structure

{ "verdict": "PASS | FAIL | INDETERMINATE", "receipt_id": "sha256:...", "signature": "...", "confidence": 1.0 }


Endpoints

https://defaultverifier.com/settlement-witness

Public keys: https://defaultverifier.com/.well-known/sar-keys.json


Agent Identity (Optional)

Format:

0xWalletAddress:agent-name

Used for:

  • attribution
  • reputation scoring

Data Handling

  • only submit structured evaluation data
  • never include secrets or private data
  • submit minimal required information

Verification Model

  • stateless
  • deterministic
  • independently verifiable
  • signature-based (Ed25519)

What This Is Not

  • not a payment system
  • not data storage
  • not enforcement layer

What This Is

  • deterministic verifier
  • receipt generator
  • proof system
  • reputation input layer

Outcome

Agents can:

  • prove task completion
  • generate verifiable records
  • build reputation over time

Keywords

verification, trust, receipts, ai-agents, validation, reputation

What ships with it: 1 file

2.5 KB alongside SKILL.md

Keep looking

Skills are one crate of 326,512. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.