Manual

Connects directly to a running Helmut4 instance via a server URL (HTTP or HTTPS) and a bearer token. The container must have network access to the Helmut4 server.

Imports streams, stream designs, store variables, preferences, metadata, metadata sets, profiles, tasks, cron jobs, users, groups, watchfolders, access presets, and product event catalogs in real time.

Imports from a local backup without a running Helmut4 system. Supported formats:

• Uncompressed mongodump directory — output of mongodump without archiving, containing database sub-directories with .bson files. Detected and labelled [DB] automatically.
• .zip / .tar.gz — a mongodump directory packed as a standard archive.
• .archive.gz — a MongoDB 8.x native archive produced with mongodump --archive.

Use the Backup Directory field to enter a container path and click the scan button (↻). The dropdown lists all eligible files and directories sorted by last-modified date, newest first. Uncompressed mongodump directories are labelled [DB], generic directories [DIR].

The selected path must be accessible inside the container — typically a bind-mounted volume such as /backup.

Defines a cron expression for recurring automatic imports. Available for both live and backup sources. Disabled by default; must be manually enabled after configuring the source.

• Live imports — connects to the configured Helmut4 server at each scheduled time.
• Backup imports — automatically selects the most recently modified file or directory from the backup scan directory. No manual selection is needed at run time.

Safety: changing the import source (live ↔ backup) immediately unchecks the schedule toggle and forces sync_enabled = false on save, so a scheduled run cannot execute against the wrong source without explicit re-activation.

Each scheduled run overwrites the current database — identical to clicking Run workflow manually. Cron expressions use 5-field standard format (e.g. 0 2 * * * = daily at 02:00). The plain-language translation is shown live as you type. All schedules are interpreted in the server's local timezone.

Overview with total counts, a severity distribution chart including acknowledged-issue breakdown and comparison against the previous import, recent issues, and the most complex streams.

Searchable, filterable list of all detected reference issues. Filter by severity level or search term. Issues can be acknowledged (hidden by default) to reduce noise after review — acknowledgements persist across workflow re-runs.

Diagnostic sub-page (accessible via the Issues sidebar item) that correlates related issues into actionable groups:

• Case mismatch — variables or metadata keys used in streams under one capitalisation but defined under another. Suggests renaming the definition to match stream usage.
• Compatibility — all Compatibility findings grouped for review with node resolution and stream deep-links.
• Deprecated — all deprecated-node findings with recommendations.
• Temp variable orphans — Read/Write Pairs, Read Without Write, and Write Without Read, with deep-links to the exact stream and node.

Each section can be exported as a Markdown report via Export Markdown.

Catalog of store variables, Helmut variables, encrypted variables, and temp stream variables. Shows type, name, status, definition source, value/expression, and usage counts. Expandable detail reveals direction, dynamic flag, and which streams read and write each variable.

Project and job metadata catalog. Shows field type, allowed values, metadata set assignments, readers, and setters. Expandable detail reveals status, mandatory/readonly/preresolve flags, and stream usage.

Lists metadata set and group definitions with their field assignments. Links to individual set detail pages.

Maps Helmut4 profiles to their main stream, pre-stream, metadata set, and product event. Expandable detail shows profile ID, priority, hidden flag, node filters, and event/host/product/tag configuration for both pre and main stream.

Catalog of all streams with complexity scores, node counts, issue counts, and reference counts. Expandable detail shows tags, reference type breakdown, and HTTP/command/JavaScript complexity sub-scores. Issue rows with node references deep-link directly to the node within the stream detail page.

All configured watchfolder definitions. Shows name, path, enabled state, recursive flag, and interval. Expandable detail shows the linked profile and any cron schedules that trigger the watchfolder.

Helmut4 task definitions. Shows name, linked profile, filter mode, and creator. Expandable detail shows the profile link and filter configuration.

All cron schedule entries sorted by type then name. Shows name, type (TASK, WATCHFOLDER, ACTIVE_DIRECTORY, …), cron expression, and resolved target. Expandable detail includes:

• Schedule — plain-language translation of the cron expression. Seconds shown only when non-zero.
• Expression breakdown — visual field-by-field grid (Second / Minute / Hour / Day / Month / Weekday).
• Target resolution — watchfolder entries link to the watchfolder by name; Active Directory entries show the CN group name and full AD configuration (groups, access preset, role, products, actions).

Standalone stream analysis and upgrade tool. Works without a database import. Paste a Helmut4 stream exchange key (plain JSON or base64-encoded) and select a reference version.

• Inspect — reports deprecated nodes, version mismatches, unknown node classes, enum option issues, and unknown triggers. Extracts all variable and metadata references. Shows a compatibility score and a Blocked state when the trigger is unknown.
• Upgrade — replaces deprecated and unsupported nodes with the best available equivalent for the target version. Parameter values are migrated by variable name; operator keys, canvas positions, and stream links are preserved. Old %variable% syntax is converted to {variable} across all nodes. Returns a migration report and a new base64-encoded exchange key ready to paste back into the stream designer.

A stream can also be opened directly from its detail page via the compatibility metric box.

Standalone stream layout preview and alignment tool. It focuses on canvas positioning rather than compatibility analysis, and is available at /node-conductor.

• Render preview — parses a Helmut4 stream exchange key (plain JSON or base64-encoded) and shows the current node layout with zoom and pan controls.
• Conduct Alignment — runs the current alignment engine and returns an aligned exchange key that can be copied back into the Helmut4 stream designer.

The alignment selector offers spaghetti_solver as the default layout engine and stream_surgeon as a built-in style-aware alternate. Node Conductor requires an active license, so it is intended for trusted users who need to inspect or normalize stream layouts before import.

Interactive vis-network reference graph covering streams, variables, metadata, profiles, tasks, cron schedules, and products. Supports Light/Dark theme, Dynamic Layout toggle, and a search field with per-node focus and match navigation (◀ / ▶, keyboard shortcuts: Enter, Arrow Up/Down, Escape).

Catalog of all individual stream nodes across the system. Shows node type, title, stream (linked), event, host, and deprecated state. Expandable detail reveals stream context, reference count, summary, references, and old wildcards.

Wildcard reference grouped by token — one row per unique wildcard regardless of stream count. Shows category, syntax, supported events/hosts, description, and a linked list of every stream that uses it. Old %percent% syntax is flagged with a badge.

Direct paginated access to all generated SQLite tables for deeper inspection of raw reference data. Hidden by default — toggle with Ctrl+Shift+T.

Interactive Swagger UI with the full OpenAPI specification for all Helmut4 Scout API endpoints. Accessible without a license.

Something referenced by a stream or configuration object is missing or cannot be resolved — for example a variable used in a node that has no definition in the system.

Deprecated stream nodes, malformed reference wrappers, or other constructs that are still functional but should be updated. The Issue Resolver Deprecated section groups these with recommendations.

Nodes using the legacy %variable.name% syntax instead of the current {variable.name} form. The Stream Workshop upgrade tool converts these automatically.

Findings generated by the Compatibility Checkup against a selected reference version. Detected conditions:

• Node not in reference — the node class is absent from the selected version. When a versioned class (e.g. ActionV5) is used but the reference only provides up to V3, the exact version gap is reported.
• Deprecated node — the node class exists in the reference but is flagged as deprecated for that version.
• Parameter value not in reference — a configured property value is not in the allowed enum list for the target version.
• Design enum values not in reference — the stream design configures option values absent from the selected version.
• Design enum options missing — the reference defines options not present in the design's configured list.
• Wildcard not in reference — a wildcard used in the stream is not in the selected version's catalog.

Potentially risky setup — for example read-only temp variables or disabled stream links.

Configured objects that appear unused or disconnected from the analysed setup.

References whose final value is resolved at runtime from stream context, user context, project metadata, node results, or string functions. Documented as detected; require manual review since their value cannot be statically determined.

Low-risk findings, often useful for cleanup or documentation purposes.

helmut_reference.sqlite — the main database used by all viewer pages. Rebuilt from scratch on each workflow run.

Documentation-friendly reference files for profiles, metadata, store variables, Helmut variables, encrypted variables, and temp stream variables. Downloadable from the Artifacts page.

helmut_reference_graph.html — the interactive vis-network relationship graph, served via the Graph view.

Static graph summaries generated when graph generation is enabled and Graphviz is available in the container. Downloadable from the Artifacts page.

issue_snapshot.json and issue_snapshot_previous.json record issue counts after each successful workflow run, powering the Dashboard severity comparison chart. Both files are removed when the database is dropped.