Changelog

The changelog tracks notable updates to the DelegateZero platform.

Changes that affect API behavior, request or response formats, or safety guarantees will be documented here. Breaking changes will be announced in advance whenever possible.

v1.5.0 — Chat, Auto-Reply Delay, and Profile Context

• Chat enhancements - The chat interface now classifies every message into one of four intents: decision testing, context exploration, context creation, and help. Decision tests run your scenario through the full pipeline in dry-run mode and return the outcome, confidence, response, and reasoning. Context exploration searches your stored knowledge using relevance scoring and highlights gaps. Context creation drafts a structured entry from natural language and asks for confirmation before saving. Help responses are tailored to your current context state.
• Conversation starters - Chat generates dynamic starters based on your existing context - entity queries, policy stress-tests, template lookups - so you can start testing immediately. New workspaces with no context see onboarding-oriented prompts instead.
• Multi-turn conversations - Chat now maintains conversation history within a session. Follow-up messages are classified with prior context, so you can refine decision tests or build on context queries naturally.
• Auto-reply delay - When auto-reply is enabled, you can now configure a delay before sending (5 minutes to 1 day). Gives you time to review and cancel. Defaults to 10 minutes. Configurable at workspace and profile level.
• Profile context sharpening - Context saved through context-building questions and other feedback loops now respects profile assignments, routing entries to the correct profile's context scope.
• Public audit page improvements - The shareable audit page has been redesigned for better readability with clearer formatting of reasoning, confidence, and response sections.

v1.4.0 — Scan Controls, Profile Intelligence, and Decision Quality

• Scan Settings - A new settings page for controlling proactive scan behavior. Configure scan sensitivity (conservative, balanced, liberal) to adjust how strictly incoming items are classified. Define a scan identity to give the classifier context about who you are and what you handle. Add excluded senders to block specific addresses or domains from scans. Toggle auto-reply and save-draft-emails at the workspace level.
• Profile overrides - Profiles can now override workspace-level scan settings including auto-reply, save-draft-emails, and scan identity. Integrations can be assigned to specific profiles so incoming items are routed through the right judgment context automatically. Profiles also support inheritance and versioned snapshots.
• Suggestion response field - The decision response object now includes a suggestion field: a single, specific recommendation generated after the decision is evaluated. Points forward only - never comments on the current result. Empty string when nothing meaningful to surface. Example: "Adding an entity entry for Jane's company would let DelegateZero tailor tone and detail level to your existing relationship history with this client."
• Extended reason field - The reason field now includes gap identification and stale entry warnings in natural prose, giving clearer insight into what context influenced the decision and what was missing.
• Decision quality feedback - After reviewing a decision, you can now submit structured feedback: whether DelegateZero should have escalated, drafted, or executed differently, and a quality rating (good, needs work, poor) with optional notes and an ideal response. Feedback is written back as high-signal context for future decisions in the same category.
• Dashboard metrics - The dashboard now shows autonomy rate, override rate, quality score, and total decisions alongside a 30-day autonomy trend projection, confidence distribution histogram, context usage breakdown by type, calibration patterns from overrides, and an entity pulse card highlighting your most-leveraged entity context.
• Processing time - Decision audit pages now display processing time from request creation to completion, making it easier to monitor pipeline performance.

v1.3.0 — Observability, Coaching, and API Improvements

• Decision Debt - A new dashboard view that surfaces open loops, stale policies, and dropped corrections before they compound into risk. Four debt types are tracked: unresolved escalations, stale policies, entity drift, and correction orphans. Includes a weekly autopsy report digest. Available on Operator and above.
• Judgment Coaching Mode - When DelegateZero detects it is operating outside its confidence range in a category, it can enter a time-limited coaching window where you provide feedback on live decisions. Your answers are written back as high-signal context and the window closes automatically once calibration is restored.
• Accountability Dashboard - A consolidated view of decision consistency over time: autonomous rate, escalation rate, average confidence, and override patterns by category - all compared period over period. Useful for understanding whether your context is improving or degrading.
• Context-Building Notifications - After every decision, DelegateZero analyses the outcome for missing or thin context and generates targeted questions to help you fill the gaps. Questions appear in the dashboard and, optionally, in email (real-time or weekly digest). Auto-suppressed per domain once confidence and escalation rates are stable. Available on all plans.
• Conflict detection - When a new context entry is saved, DelegateZero scans for logical conflicts with existing entries (contradictory policies, overlapping precedents, etc.) and surfaces them before they silently affect decisions. Conflicts are flagged inline with the affected entries identified.
• Background context ingestion - Context entries are now indexed asynchronously. Saving a large source or policy no longer blocks the UI - the entry appears immediately with an "Indexing..." indicator and becomes active once processing completes.
• Email notifications - Configurable email delivery for context-building questions (real-time and weekly digest) and the weekly decision summary. Managed in Settings > Notifications.
• Pinned context via API - The decisions endpoint now accepts a context_ids parameter. When provided, DelegateZero skips the relevance search and consults only those entries. Useful when your integration already knows which policies or entities apply - produces faster, more deterministic results and a cleaner audit trail.
• Context ID column - Context IDs are now visible and copyable on every context page (Policies, Entities, Playbooks, etc.), making it straightforward to build context_ids allow-lists for API requests.

v1.2.0 — Proactive Decision Surfacing

• Gmail integration - Connect your inbox and DelegateZero will scan for incoming emails that require a decision or approval. Qualifying messages are classified, run through the decision pipeline, and waiting in your queue with a drafted response before you open the app.
• Linear integration - Connect your Linear workspace and DelegateZero will identify new issues that require executive judgment before work can proceed. Issues are scanned every 15 minutes.
• Proactive queue - A new section at the top of your Queue surfaces all proactively identified items. The queue header tells you at a glance how many are waiting, how many are already drafted, and how many need your direct input. Each item links to the full decision for review.
• Classification pipeline - Incoming messages and issues pass through a lightweight classification step before entering the decision pipeline. Newsletters, automated notifications, and routine messages are filtered out. Only items with a concrete decision point are surfaced.

v1.1.0 — Memory, Relationship History, and Context Intelligence

• Memory - New system-authored context type. DelegateZero now writes a Memory record after every decision, capturing the request, outcome, confidence, and reasoning. Correction events (user overrides) are recorded as high-signal Memory records that directly influence future decisions of the same type.
• Relationship Memory on Entities - Entity entries are no longer static. When a decision involving a known entity reveals net-new signal, the entity entry is automatically rewritten as a tighter, more current narrative. Entity context now reflects relationship history, not just an initial profile.
• Context Freshness - All context entries now carry freshness metadata. Entries that exceed type-specific staleness thresholds are flagged as potentially stale. Staleness flags appear in audit logs with entry name and age whenever a stale entry influenced a decision.
• Confidence Autopsy - Weekly digest surfacing patterns across escalations and identifying the context changes most likely to resolve them. Available on Operator and above.
• Decision Simulation - Batch replay historical decisions against proposed context changes before committing them. Available on Operator and above.
• Judgment Profiles - Portable, exportable snapshots of a user's decision-making compiled from Memory, Policies, and Precedents. Available on Team and above.
• Delegation Chains - Multi-user decision hierarchies where policies flow downstream and overrides are scoped per layer. Available on Team and above.
• Updated plans - Solo ($49), Operator ($149), Team ($399), Scale ($999). See Plans for details.

v1.0.0 — Initial release

The initial public release of DelegateZero focuses on safe, explainable decision delegation.

• Decision API for requesting judgments and actions on your behalf
• Flexible request object with request, context, and instructions
• Structured response object with confidence scoring and audit links
• Confidence-based execution, drafting, and escalation
• Context ingestion via dashboard and /context API endpoint
• Support for policies, precedents, entities, playbooks, templates, and sources
• Immutable audit logs for every decision
• Webhook notifications for decision creation and escalation
• Dashboard-based configuration for safety and attribution
• Chat interface for request/response testing

Future releases may expand execution capabilities, integrations, and observability.