App map across web and API.

A single-screen consultant cockpit that keeps issue intake, status, and navigation visible.

• Sticky top bar with brand, theme, plan, AI, and team status.
• Three saved themes: Classic, Black, and Henry.
• Quick jumps, workspace pulse, and metric cards on home.
• Guest-safe landing state plus authenticated account controls.
• Client-side persistence for auth and theme.
• Error banner when the API is unavailable.

The core support engine that turns issue text, screenshot hints, and context into next actions.

• Live AI diagnosis with provider-aware reasoning metadata and citations.
• Online-only search mode across live provider reasoning, custom content, ingested knowledge, saved cases, and solutions.
• Merged results from AI guidance, custom content, ingested knowledge, saved cases, and solutions.
• Reference hub plus T-code, table, message, config-path, and field search.
• Transport impact analysis for STMS, SE09, and SE10.
• Auth cockpit for SU53, STAUTHTRACE, ST01, PFCG, catalogs, and backend checks.
• IDoc and interface recovery for WE02, WE05, BD87, SM58, and qRFC.
• OSS Note and KBA search from symptoms, release, and component clues.
• Master data validation for plant, company code, sales area, purchasing org, and partner or material chains.
• Job, output, and spool triage for SM37 and SP01.
• RCA assistant with causes, evidence, control points, and narrative.
• Guided troubleshooting plus saved yes/no wizard sessions.
• Screenshot triage with AI vision or metadata fallback.
• Recommended test steps, escalation paths, module summaries, and local search history.

Operational tooling for turning diagnosis into owned work, reusable fixes, and structured documentation.

• Save issue matches as cases from the dashboard.
• Case lifecycle with status, assignee, notes, summary, comments, and activity.
• Resolution feedback that can promote useful fixes into draft solutions.
• Reusable solution library for organization-specific playbooks.
• Promote successful candidates into formal solutions.
• Ten built-in document templates for RCA, specs, notes, KT, SOP, and test cases.
• Cutover, hypercare, and period-close controls for ownership, evidence, rollback, and repeat incidents.
• Generated markdown documents saved in runtime storage and visible on the dashboard.

Features that expand the system beyond seed data and train consultants over time.

• Custom knowledge articles with root cause, context, references, solution steps, and keywords.
• Enterprise text and file ingestion with chunking, previews, tags, and searchable runtime storage.
• Facet summaries for modules, process areas, error types, transactions, and integration topics.
• Document views for ingested knowledge with chunk excerpts.
• Quiz, flashcard, and mock-issue learning sessions.
• Answer scoring, mastery tracking, and per-module learning analytics.
• Knowledge sync that reloads the retriever and logs sync jobs.

Shared-work features for support teams that need assignment, collaboration, and expert routing.

• Registration, login, session-backed auth, and current-user bootstrap.
• Organization-scoped user lists and manager-created users.
• Team creation, membership management, and team-scoped access.
• Seed experts with module specialties and SLA hints.
• Escalations with lifecycle updates, assignments, comments, and history.
• Notifications for welcome, plans, team events, assignments, generated content, and sync completion.

Commercial controls, usage visibility, and outbound workflow connectivity.

• Plan catalog with free, standard, pro, ultimate, credits pack, module pack, and enterprise options.
• Active subscription handling, credits usage, and manager-level administration.
• Simulated checkout and payment history for local testing.
• Analytics for searches, cases, documents, learning, escalations, knowledge gaps, time saved, and mode usage.
• Team analytics for shared activity and knowledge coverage gaps.
• Outbound integrations for generic, Slack, and Teams webhook styles.
• Integration test delivery, echo endpoint, and delivery logs.

Provider-aware features and runtime architecture that support future upgrades without changing the UI.

• OpenAI and OpenRouter hooks for reasoning, web search, file search, image understanding, and extraction when configured.
• OpenAI vector-store upload path for ingested documents when OpenAI mode is active.
• PostgreSQL-backed runtime persistence keeps the workspace state in a live database when DATABASE_URL is configured.
• Workspace bootstrap that aggregates health, modules, analytics, content, teams, notifications, and subscriptions.
• Single FastAPI runtime that serves both the Jinja2 web workspace and the API compatibility paths.