Replay
11 essays tagged with Replay.
Proof-Carrying Agent Runtime: A Bounded Architecture and Specification
A research-grounded runtime proposal with explicit proof obligations, an effect-boundary state machine, a conditional safety argument, and failure tests.
Multi-Agent Systems Need Replay Before More Agents
Multi-agent systems multiply coordination, cost, and causal ambiguity. Prove one agent is replayable before adding lanes and handoffs.
Harness Improvement Loops Need Replayable Environments
Why harness improvement needs replayable episodes, bounded mutations, scorecards, source closure, and promotion gates.
Agentic AI Systems Before and After ContextOS
A concrete before-and-after guide to agentic AI: what context, authority, evidence, replay, and release controls change in a refund run—and what they cannot guarantee.
Build an AI Agent Replay Test with the Context Pack Compiler
Run four TypeScript tests for repeatable context compilation, tool permissions, approval thresholds, and missing evidence. Includes code and replay limitations.
End-to-End Refund: How 12 Primitives Compose in One Production Run
A single refund run traced through 12 ContextOS primitives, from invokeAgent envelope to byte-equal replay.
Context Graphs: Decision Lineage as a System of Record
How hash-chained DecisionRecords turn execution-time context into a queryable lineage graph for why an agent acted.
From Operator Correction to Released StrategyRule: The Improvement Loop, Coded
How one operator correction becomes a reviewed, replayed, versioned StrategyRule that prevents repeat agent failures.
Replay Is the Real Audit Log
Why "we have logs" is not an audit story, and what a hash-chained Decision Record plus canonical replay actually buys you when an incident hits.
Beyond Prompts: The Architecture of Trust for Agentic AI
Building a governed decision runtime across Intelligence, Context, Decision, Action, and Trust — with evaluator scoring, approval tiers, and replay-bound audit.
Context Engineering in Production
Why most agent failures are not model failures — they are context failures — and what changes when context becomes a versioned, testable, replayable contract.