Changelog

Agents, Investigate, Audit, Routing, and Budgets now have more focused views, clearer language, and faster ways to find the records that need attention.

Permit activity now distinguishes previews, approvals that expired before execution, closed budgets, and usage still awaiting measurement, giving teams a more accurate operational picture.

Eligible customers can now apply promotion codes directly during hosted Production checkout.

Organization and team features now recognize the eligible plan held by the owning account, making access match purchased service more consistently.

Permit review can now be navigated by agent, child agent, action, and ungrouped activity, with counts and filters that remain aligned.

Unchanged connector tool definitions now renew automatically during governed calls, while setup-related holds clearly identify the action needed to resolve them.

Evidence-bundle verification now confirms that the signing key comes from an approved trust source and remains valid for the bundle being checked.

Supported generation requests now reach active policy evaluation more consistently, while approval eligibility and product guidance align across review surfaces.

Usage reporting now distinguishes verified costs from estimates more clearly, with broader pricing coverage and more dependable spending-limit decisions.

API keys now show their type, expiration, recent use, and current status more clearly, with safer creation and replacement guidance.

Compliance exports now prioritize a complete proof pack, making signed evidence easier to download and verify independently.

Governed workflows now handle interruption and completion more reliably, while policy and operational views report activity more consistently.

Starter setup now provides clearer request examples, more useful connector guidance, and a simpler handoff after the first governance proof.

Go SDK 0.5.0 expands support for current connector and governance evidence workflows, with updated guidance for existing integrations.

Integrity verification is now more deliberate and easier to interpret, with timestamped results and more dependable dashboard behavior.

Additional provider SDK workflows are supported, while connector setup and maintenance now provide clearer operational guidance.

Governance evidence is easier to review as a connected journey, with more consistent authority validation across decision workflows.

Keel's supported SDKs and infrastructure integrations now align more closely with current API capabilities and operational workflows.

Policy review is more consistent across authoring modes, while clearer decisions and reliability improvements make governed workflows easier to operate.

Guided setup now takes new teams from project creation through private credential verification and a reviewable first permit.

Real-time AI workflows now support additional model options, with clearer guidance about which models are available for governed use.

Permit details now support faster operational review while keeping complete technical evidence available when teams need it.

Policy authoring now separates saved changes from activation, and trusted action availability documentation is generated from the API contract.

Voice and telephony workflows now have broader governance coverage for operational actions and reviewable evidence.

Keel added public visibility for its CSA STAR for AI Level 1 self-assessment alongside existing trust materials.

Permit evidence exports and public control-framework materials are easier for teams to inspect and use during governance review.

Policy and review workflows now give teams a clearer way to manage agent rules, review pending work, and preserve governance evidence across operational workflows.

Authority workflows now give teams a more organized way to review agent status, manage follow-up actions, and retain evidence for governance review.

Policy authoring, action controls, and permit review flows now give teams more useful guardrails and evidence.

Governance checks now cover authority, policy state, and budget-controlled work more consistently.

Keel now gives teams a clearer view of which agents, workers, and delegated controls are active across a project.

Signed compliance evidence is easier to move, verify, and review across governed payment workflows.

Keel improved realtime dashboard updates, permit decision evidence, and session handling for governance workflows.

Cost permit visibility, budget-aware enforcement, and API documentation now line up more closely for operators.

Keel improved evidence quality, verification consistency, and reliability for governed payment workflows.

Project dashboards, compliance export access, and operational views are easier to navigate and rely on.

Keel added support for governing Stripe Machine Payments Protocol payment flows, including spend request validation, challenge handling, execution evidence, and latency visibility.

Policy enforcement behaves more consistently across governed request paths, including actions, connectors, routing decisions, and output limits.

Keel expanded governed action support across more tool and connector workflows, with richer action evidence and clearer approval, cost, and connector visibility.

Production-tier behavior, dashboard views, and documentation now align more closely, with broader evidence-pack support for checkpoints and exports.

Keel improved how exported evidence and integrity artifacts are verified, with clearer failure handling and more dependable validation paths.

Related decisions, permits, and evidence are easier to group around a single workflow, with updated SDK support for passing workflow context consistently.

Expanded dashboard-operated MCP and connector workflows, improved governance evidence views, added compliance evidence references, and strengthened execution observability and runtime controls.

Expanded policy authoring, strengthened execution-time constraint handling, standardized throttling behavior, and improved reason codes for review, alerting, and SDK handling.

Improved Keel SDKs with easier governed execution setup, retry and timeout handling, stronger permit checks, and clearer examples across supported languages.

Added clearer platform status visibility, lightweight health checks, stronger execution management controls, and improved abuse resistance.

Aligned governed execution behavior across supported surfaces and strengthened validation, usage accounting, and audit consistency for permitted workloads.

Strengthened service credential validation and expanded audit evidence for privileged operations.

Updated API examples and reference links so published documentation more closely matches current platform behavior.

Improved usage ledger reconciliation and export verification. Customers with affected finance workflows received direct revalidation guidance.

Keel began with a simple operating principle: AI execution should be governed before work is allowed to run. From the start, Keel has focused on permit-based decisions, reviewable evidence, and controls that help teams prevent unsafe or unnecessary AI spend before execution.