What it is (1 sentence)
Open Line Protocol (OLP) is a tiny, typed JSON wire so agents exchange small graphs + telemetry instead of paragraphs.
Why graphs, not text?
Because we want auditable plans: nodes are Claims/Evidence/Steps; edges are supports/depends/updates. The bus recomputes a 5-number digest (b0, cycle_plus, x_frontier, s_over_c, depth) and a holonomy gap Δ_hol to catch self-reinforcing loops and “too-clean” rewrites.
What’s new vs. just “send JSON”?
- *Frozen schema v0.1* (back-compat promise)
- *Guards* (rejects self-licking loops, silent objection deletion, order-debt spikes)
- *Receipts* (machine-verifiable JSON, schema-checked in CI)
- *Telemetry* (phi_sem/topo, kappa_eff, commutator) so agents can auto-throttle
1. A structured data format for representing arguments/reasoning as graphs
2. Validation rules to prevent circular reasoning and inconsistencies
3. A protocol for AI agents to exchange these structured argument graphs
Author here—happy to answer questions.
What it is (1 sentence) Open Line Protocol (OLP) is a tiny, typed JSON wire so agents exchange small graphs + telemetry instead of paragraphs.
Why graphs, not text? Because we want auditable plans: nodes are Claims/Evidence/Steps; edges are supports/depends/updates. The bus recomputes a 5-number digest (b0, cycle_plus, x_frontier, s_over_c, depth) and a holonomy gap Δ_hol to catch self-reinforcing loops and “too-clean” rewrites.
What’s new vs. just “send JSON”? - *Frozen schema v0.1* (back-compat promise) - *Guards* (rejects self-licking loops, silent objection deletion, order-debt spikes) - *Receipts* (machine-verifiable JSON, schema-checked in CI) - *Telemetry* (phi_sem/topo, kappa_eff, commutator) so agents can auto-throttle
Try it fast (no infra): • Readme + schema/digest examples: https://github.com/terryncew/openline-core • Live receipts hub (static): https://terryncew.github.io/openline-hub/
What feedback is most useful? 1) Field naming in the wire 2) Guard thresholds 3) Which adapter you’d want first (WebSocket, store).
What dictionary was murdered to create these terms?
1. A structured data format for representing arguments/reasoning as graphs 2. Validation rules to prevent circular reasoning and inconsistencies 3. A protocol for AI agents to exchange these structured argument graphs
Is this a reasonable description?