Portable replay records for AI-agent runs.
A public schema and reference validator for packaging AI-agent runs into portable evidence artifacts. A replay bundle records the task frame, action proposals, policy decisions, evaluation traces, blocked actions, authority records, reliance records, final output, validation report, redaction metadata, and signature metadata.
Agent runs often span prompts, tools, APIs, databases, files, workflow systems, and policy checks. After an incident or review request, teams may not have a single portable artifact that reconstructs what happened.
Agent activity lands in prompt logs, tool logs, API logs, and workflow systems, with no single place that reconstructs the run.
Each agent framework records tool calls differently, so evidence cannot move cleanly between systems or teams.
When an action is denied, the denial often never reaches durable storage, leaving gaps in the record.
Which file, API, or model output the agent relied on is rarely captured in a form that survives the run.
Ad-hoc exports carry no consistent proof of integrity, no redaction record, and no validation result.
Without one portable artifact, reconstructing an agent run after the fact depends on stitching together systems that may have moved on.
The task context the run began from, including inputs and configuration.
Every action the agent proposed during the run, before any decision.
The allow, block, or escalate outcome recorded for each proposed action.
The ordered rule path behind each decision, showing which rule matched.
Structured records of the actions that were denied, and the reasons why.
Who or what authorized an action, and under what scope.
The tools, files, APIs, and outputs the agent relied on across the run.
The result the agent produced once the run completed.
The semantic consistency check across the bundle's records.
What was redacted from the exported copy, while preserving structure.
Optional integrity metadata so the bundle can be checked for modification.
Agent Replay Bundle is the Replay layer. It packages the output of runtime control into a portable artifact that can be validated, redacted, signed, stored, or transformed into a business-facing evidence pack.
Need a single portable artifact that reconstructs an agent run during investigation.
Need consistent, framework-independent evidence of what an agent run contained.
Need a common export format their control plane and tools can emit and consume.
Need to inspect blocked actions, reliance, and decisions without access to live systems.
Need a schema and validator to package runs into replayable evidence.