Name the core abstraction and its failure modes.
Maple AI OS Course
WorldLine receipts and proof-carrying work
How Maple uses receipts to make autonomous work attributable, replayable, and verifiable.
Learning Objectives
What this lesson should make precise
Translate the concept into a Maple/Hermon proposal contract.
Define at least one evaluation case that can fail the model safely.
Tutorial Flow
How this lesson becomes a demo and training target
Each tutorial is written as a user education path and a model-improvement artifact. The diagram shows how the idea moves into a lab, a typed contract, an eval gate, and then a Hermon/MapleAI demo route.
Concept
Receipt anatomy
Applied Lab
Applied lab: WorldLine receipts and proof-carrying work
Output Contract
intent_hash, actor, policy_version, authorization
Eval Gate
Does the answer separate model proposal from deterministic execution?
Demo Route
Maple AI OS training and public model checks
01
Receipt anatomy
A receipt should contain intent, actor, policy version, input references, authorization result, execution result, model route, and rollback state. It does not need to expose secrets to be useful.
- Hash sensitive inputs.
- Store policy version.
- Record model and adapter.
02
Replay and audit
Receipts let operators ask whether a result was allowed by the policy at the time, whether the model version was approved, and whether rollback is still possible.
- Replay decisions.
- Compare expected effects.
- Detect policy drift.
03
Proof of Contribution
Once work is receipt-backed, Maple can reward verified contributions rather than noisy activity. This is the path from agent logs to accountable distributed AI labor.
- Reward verified outcomes.
- Slash dishonest verification.
- Preserve provenance.
Lab
Applied lab: WorldLine receipts and proof-carrying work
Write the receipt fields for a model adapter rotation with canary rollout, rollback, and post-deployment health checks.
Expected result
- A typed JSON-style proposal rather than free-form advice.
- Clear authority boundaries and denied operations.
- A test or rubric that decides whether the proposal is deployable.
Evaluation
How Maple would grade this work
Rubric
- Does the answer expose assumptions instead of hiding them?
- Does the answer separate model proposal from deterministic execution?
- Does the answer produce artifacts that can be tested, reviewed, and rolled back?
Output contract
intent_hash, actor, policy_version, authorization, execution_result, rollback_stateUse this lesson as training direction
A strong lesson gives users a mental model and gives Hermon a sharper target for examples, probes, and demo prompts.