Proof Reporttrust artifact
evidence-first autonomyEvidence before autonomy scales.
A Proof Report turns autonomous execution into reviewable operational evidence before PR, merge, deployment, escalation, or rollback.
Proof Report/v1/proof-packets/sample
TraceWorkGraph event cursor, runtime receipts, tool calls, actor identities, and capability refs.
6/6 required lanes present Checkstypecheck, lint, build, diff hygiene, and capability-contract assertions.
validation gate required Artifactspatches, logs, screenshots, generated files, capability readbacks, and proof report refs.
artifact rail visible Riskbranch race, path ownership, external side effects, and deployment blast radius.
human review required Rollbackbaseline ref, revert command, supersession marker, replay cursor, recovery owner.
rollback path pinned Approval Queue/v1/governance/decisions/sample
sensitive actionProof Report must be accepted before external or sensitive action
human_reviewerapproval_required External side effectNo external send, production mutation, or customer-visible change from the console
owner onlyblocked Runtime escalationSpecialized execution lanes require explicit policy and spend scope
governance_corereview DeploymentRelease handoff needs health checks and rollback review
release_handoffhold Destructive shellDenied until explicit command, target, and rollback are present
toolmeshblocked Rollback Inspector/v1/events/feed
origin/mainbaselinelast fetched remote before autonomous work
anchorworkgraph://verified-prstate checkpointobjective, leases, runtime choices, policy decisions
readyreview://report/sampleevidence bundlechecks, logs, trace, artifacts, reviewer checklist
compilingartifact://patchreversible diffsupersede or revert before sensitive action
reviewevent://replayreplay cursorresume from last verified event if runtime fails
waitingTool Permissions View/v1/tool-registry/capabilities
GitHubchecks/read by default; PR/write gatedWorkGraph lease + Proof Report
approval_required Shellsandboxed commands only; destructive deniedExecution receipt + side-effect summary
leased Browserobserved navigation, screenshots, console logsExecution Chamber artifact
leased Protected resourcesSensitive systems exposed by contractToolMesh capability + denial receipt
review SkillsExecution Host/OpenAI/agent skills wrapped as governed actionspolicy scope + review ref
ready EvaluationsQuality reviews feed the release decisionquality gate + risk note
approval_required review gateObjective
Summary, repo, branch, actor, risk class, and acceptance criteria are frozen before execution.
review gateReceipts
Every runtime/tool action records input, output, side effects, artifacts, and policy decision.
review gateVerification
Tests, evals, screenshots, build logs, review notes, and capability readbacks are linked.
review gateApproval
Human can approve, reject, narrow, retry, stop, or rollback with the review in view.
review gateLineage
Commit, artifact, runtime version, source repo, and rollback target remain traceable.
fieldobjective
Required for governed handoff and auditability.
fieldplan
Required for governed handoff and auditability.
fieldexecution trace
Required for governed handoff and auditability.
fieldartifacts
Required for governed handoff and auditability.
fieldtests/evals
Required for governed handoff and auditability.
fieldreview notes
Required for governed handoff and auditability.
fieldrisks
Required for governed handoff and auditability.
fieldapprovals
Required for governed handoff and auditability.
fieldrollback
Required for governed handoff and auditability.
fieldlineage
Required for governed handoff and auditability.
fieldhandoff
Required for governed handoff and auditability.