Numonic
Server Details
Search, organize, and publish AI-generated images and video with provenance and lineage.
- Status
- Healthy
- Uptime
- 100.0% over the last 24 hours
- Last Tested
- Transport
- Streamable HTTP · MCP 2025-03-26
- URL
- ChatGPT
- –
- Claude
- –
TDQS
Scored across 23 tools
Most tools target distinct objects (assets, collections, models, prompts, pipelines, rules, analytics) and descriptions proactively cross-reference each other. The real overlaps are GetAssetLineage vs GetMidjourneyEvolutionChain (the latter is explicitly kept as a legacy Midjourney-parameter-diff tool) and GetAssetPublicUrl vs GetCollectionPublicUrls, both of which take some reading to separate. GetAssetDetails vs GetAssetModels is also close but clarified in-text.
Every tool is PascalCase VerbNoun with a predictable verb set (Get*, List*, Render*, Summarize*), so the pattern is immediately readable. The singular/plural and scope suffixes (GetAssetPublicUrl vs GetCollectionPublicUrls, GetModelDetails vs GetModelVersions) are used consistently to signal the resource and scope.
23 tools spans many sub-domains (assets, lineage, models, prompts, collections, pipelines, rules, search analytics, storage, experiments), so no single domain feels over-stuffed, but the total is on the heavy side of the 16-25 band. A handful (GetExperimentSummary, ListExportPresets, GetTenantStorageDetails) feel peripheral to the core asset/lineage purpose.
The surface is read-only in a domain that clearly needs writes: no asset upload/create, no CreateUploadLink (despite GetUploadLinkStatus describing how to read one back), no AddToCollection/UpdateAsset/PublishAsset, no CreateCollection, no CreateRule, and no RunSavedPipeline or CreatePipeline despite ListPipelines/ListPipelineTemplates pointing at them. Multiple tool descriptions route the agent to operations that do not exist here, producing dead ends, and there is no search tool even though GetSearchAnalytics reports on search.
Publisher details
- Operator
- Numonic Labs Ltd. · Publisher source
- Operator website
- https://www.numonic.ai/
- Vendor relationship
- First-party
- Documentation
- https://www.numonic.ai/docs/api/mcp
- Privacy policy
- Not provided
- Trust center
- https://www.numonic.ai/trust
- Restrictions
- Works with free and paid accounts
Available Tools
23 toolsGetAnnotationsBInspect
Retrieve all annotations for a specific ComfyUI workflow node with author information.
Inputs: workflow_node_id (required), include_resolved.
GetAssetDetailsAInspect
Retrieve comprehensive details for a specific asset by its SHA-1 hash, including how it was made. Answers 'which prompt, model and workflow produced this?' directly — no download needed: prompts.positive/prompts.negative hold the prompt, models[] lists every model the workflow loaded (name, type such as checkpoint/unet/lora/vae, loader node_id), workflow.workflow_json is the full ComfyUI graph (with workflow.tool_name, node_count, custom_nodes), and generation_parameters holds sampler settings (model_name, seed, steps, cfg_scale, sampler, scheduler) when they were read from a standard KSampler node, else null (the graph still has them). provenance.lineage_source says where that lineage came from: 'extracted' (read by Numonic from the stored file; prompts.source/workflow.source are then 'extracted'), 'client_asserted' (only the uploader's claim), or 'unknown' (nothing asserted or found). provenance.asserted_parameters holds the uploader's own claim about the generator, model and settings (StoreAsset prompt_metadata.parameters_used, e.g. {generator: 'fal', model: 'fal-ai/flux/schnell'}), or null; it is client-asserted and never merged into generation_parameters. Also returns title, description, tags, collections, provenance (creation prompt, parent assets), embedding info and copyright. For Model Library records of those models (base model, trigger words, usage) call GetAssetModels. Optionally generates a short download link for the original file (valid 5 minutes; fetch it with a tool such as curl or fetch rather than retyping it), and optionally returns a small inline preview image (include_image) so the asset can be seen without fetching a URL. For a URL that does not expire, approve the asset and call PublishAsset.
Inputs: asset_h (required), tenant_id, include_image, include_download_url, +1 more.
GetAssetLineageAInspect
Read an asset's lineage as a graph, across every source tool (not only Midjourney): its ancestors and descendants as nodes and edges. Returns {schema_version, asset_h, scope, nodes[], edges[], annotations[], truncated}. Each node has source_tool and tools[] (tool, role, attribution_source) and branch_decision (null = none recorded). Each edge has parent, child, relationship_type, derivation_kind and: state (visible by default; with include_hidden also disputed = sources disagree, not_live = no active assertion), assertion_sources (who currently affirms the edge: extracted = read from the file, user_asserted = a person, client_asserted = an integration or client), match_method (how the parent was identified, e.g. content_hash, mj_job_id, c2pa_ingredient, filename, declared), disagreement and legacy_unasserted. With include_hidden each edge also lists every source's position. annotations are non-lineage links (ReferenceFor, ComponentOf, AlternativeTo) one hop from a returned node; they are never walked. scope "lineage" (default) = ancestors up and descendants down; "family" = the whole connected component, so siblings such as Midjourney grid images appear. truncated is true when max_depth or max_nodes cut the walk short. An unknown, other-tenant or hidden asset returns the empty graph. GetMidjourneyEvolutionChain remains for Midjourney parameter diffs.
Inputs: asset_h (required), scope, max_depth, max_nodes, +1 more.
GetAssetModelsAInspect
Get the Model Library records for every model used to generate a specific asset (checkpoints, UNETs, LoRAs with weights, VAEs, CLIP, upscalers), with base model, type and detection source. Scoped to your tenant: an asset your tenant does not hold returns no models. For the prompt and ComfyUI workflow graph, call GetAssetDetails.
Inputs: asset_hash (required), tenant_id.
GetAssetPublicUrlAInspect
Reverse lookup: finds if an asset is in ANY published collection and returns the public URL. Returns one of three states: (1) Asset IS published with public_url, collection_path, preset, published_at; (2) Asset NOT published but in collections: lists collection paths; (3) Asset NOT in any collection: empty collections array.
Inputs: asset_h (required), tenant_id.
GetCollectionAssetsAInspect
Get assets in a collection with membership metadata (position, role, added date). For a query-backed collection (collection_kind "query") the assets are the current matches of its saved query, newest first; include_nested and order_by do not apply.
Inputs: collection_path (required), limit, offset, order_by, +2 more.
GetCollectionPublicUrlsAInspect
Get publication status and public URLs for all assets in a collection. Returns whether the collection is published, publication metadata, and URLs for each published asset.
Inputs: collection_h (required), tenant_id.
GetCreativeSessionAInspect
Discover the "creative session" around a given asset: other assets from the same tool made within a time window of it. Midjourney assets are grouped by job enqueue time. ComfyUI and other tool-evidenced assets are grouped by recorded generation date, or by upload time when the file records none (each asset reports its time_source). Assets with no tool evidence have no session.
Inputs: asset_id (required), time_window.
GetExperimentSummaryAInspect
Get aggregated experiment results: run count, condition breakdown, outcome tag distribution, and comparison count.
Inputs: collection_h (required).
GetMidjourneyEvolutionChainAInspect
Retrieve the evolution chain (parent→child lineage) for a Midjourney asset, showing how it evolved through variations and upscales. Steps that are another member's private generations appear as placeholder nodes (redacted: true, job_type "private", no asset, prompt or command); describe them as private steps rather than inferring their content.
Inputs: asset_id (required), max_depth.
GetModelDetailsBInspect
Get full details for a specific model including metadata, version history, asset usage count, and preview count.
Inputs: model_hash (required), tenant_id.
GetModelVersionsBInspect
Get version history for a model, including training steps, quality scores, and recommended flag.
Inputs: model_hash (required), include_deleted.
GetPromptBInspect
Get a prompt by hash ID with full version history and evaluations.
Inputs: prompt_h (required), tenant_id.
GetSearchAnalyticsAInspect
Retrieve search analytics summary with health assessment (green/amber/red) for monitoring search quality. Returns zero-result rate, latency percentiles (P50/P95/P99), query-type breakdown, and top zero-result queries. Health thresholds: zero-result rate (<15% green, 15-25% amber, >25% red), P95 latency (<500ms green, 500-1000ms amber, >1000ms red).
Inputs: period, tenant_id.
GetTenantStorageDetailsAInspect
Retrieve storage usage details for a tenant including bytes used, GB used, storage limit, percentage used, and over-limit status. Helps monitor storage consumption and enforce storage quotas.
Inputs: tenant_id.
GetUploadLinkStatusAInspect
Read back an upload link created by CreateUploadLink: whether it is still open, closed or expired, the asset_h of every file uploaded through it, and, if you passed expected_sha1, which digests matched, are missing or were unexpected. Use the returned asset_h values with AddToCollection, UpdateAsset and GetAssetDetails. Poll after asking the user to upload; do not assume the files arrived.
Inputs: upload_link_id (required), tenant_id.
ListCollectionsBInspect
List collections as a hierarchical tree for organizing assets. Supports filtering by parent path and depth. Each item has collection_kind: "static" (assets filed by hand) or "query" (a saved search whose contents are whatever its query matches when read; its query is returned).
Inputs: max_depth, tenant_id, parent_path, include_asset_counts.
ListExportPresetsBInspect
List available export presets with their default options. Returns the preset configurations used by privacy-aware exports in the Numonic app and REST API (ADR-057).
Inputs: tenant_id.
ListPipelinesAInspect
List saved pipelines for the current tenant. Returns pipeline metadata including name, stage count, tags, and default execution settings. Use pipeline_definition_h from results with RunSavedPipeline to execute.
Inputs: tags.
ListPipelineTemplatesAInspect
List available pipeline templates for the current tenant. Returns system templates (available to all tenants) plus any tenant-specific templates. Use the pipeline_definition_h from results with POST /api/v1/pipelines/templates to clone a template into a new saved pipeline.
Inputs: no named parameters.
ListRulesBInspect
List automation rules for the current tenant with optional filters. Returns rule metadata including name, trigger type, enabled status, and linked pipeline.
Inputs: enabled, tenant_id, trigger_type.
RenderPromptBInspect
Render a prompt template by substituting {{variable}} placeholders with provided values.
Inputs: prompt_h (required), tenant_id, variables.
SummarizeStageCInspect
Execute a single SUMMARIZE-category pipeline stage on a set of assets. Produces aggregate data and statistics. Supported stage_type values: count, group_by, stats, histogram.
Inputs: assets (required), stage_type (required), config, dry_run.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Related MCP Connectors
Upload, generate, organize, search, and transform images, videos, and files with AI-powered tools.
Upload, generate, organize, search, and transform images, videos, and files with AI-powered tools.
AI-native digital asset management: semantic search, generative image edits, and CDN delivery.
Digital asset manager - upload, organize, search, and generate on-brand images and videos
Related MCP Servers
- AlicenseNot gradedqualityBmaintenanceEnables agentic creators to generate images, videos, and audio, run taste-based scoring, and publish through human-gated signed manifests using their own keys and local provenance.MIT
- AlicenseNot gradedqualityDmaintenanceGenerate and refine AI images/audio/video through natural conversation.412Apache 2.0
- AlicenseCqualityBmaintenanceOpen-source local-first AI media workbench with Canvas, video editor, Assets, and MCP control.38348Apache 2.0
- AlicenseNot gradedqualityBmaintenanceEnables agentic creators to generate, score, and publish image, video, and audio assets using their own API keys and local storage, keeping a content-addressed provenance ledger and selecting the best candidates against pack rubrics. Brand-specific workflows layer on top as data packs, and posts go out only through a signed, human-confirmed manifest to the user's own publishing channel or local outbox.MIT
Glama MCP Gateway
Add one secure layer between your agents and this server.