"Hugging Face" matching MCP tools:

• hub_repo_searchA

  Hugging FaceA

  Search Hugging Face repositories with a shared query interface. You can target models, datasets, spaces, or aggregate across multiple repo types in one call. Include links to repositories in your response.

  ConnectorNo auth

• hub_repo_detailsA

  Hugging FaceA

  Get details for one or more Hugging Face repos (model, dataset, or space). Auto-detects type unless specified. For datasets, use operations: overview, dataset_structure, dataset_preview. Use dataset_structure first to discover configs, splits, sizes, and schema. Use dataset_preview only when config and split are known, unless the dataset has a single config/split.

  ConnectorNo auth

• hf_whoamiA

  Hugging FaceA

  Inspect the current Hugging Face authentication context, including the account, visible organization memberships, and credential access details. Read-only and never returns credential values.

  ConnectorNo auth

• data_previewB

  New Hugging Face Spaces — AI App Demo Discovery (hfspaces)A

  ✅ No API key needed — call this now. Listing: hfspaces: New Hugging Face Spaces — AI app demos & live web UIs. Price 0.01 USDC/query (max 20 queries/session). Sample questions: What AI app demos were just created on Hugging Face Spaces?; Show the newest gradio or streamlit Spaces shipped this hour. FREE preview — no key, no payment. Try one of the sample questions now.

  ConnectorNo auth

• validate_claimA

  FacA

  "Is it true that…" / "fact check" / "verify the claim that…" / "did X really…" / "was Y actually…" / "confirm or refute" / "true or false" — natural-language claim verification against authoritative sources. Use whenever the agent needs to check whether something a user said is factually correct. Company-financial claims (revenue, net income, cash for public US companies) verify via the structured SEC EDGAR + XBRL fast path with exact percent-delta math; ANY OTHER factual claim (macro statistics, rates, prices, drug data, records) automatically falls through to the grounded pipeline — routed to the right live source, answered with verbatim evidence, then judged. Returns a verdict (confirmed / approximately_correct / refuted / inconclusive / unsupported / could_not_verify), the grounded or structured actual value with pipeworx:// citation, and reasoning. IMPORTANT for callers: could_not_verify means the check did not happen (our LLM or source failed) and carries verification_error{stage,detail} — it is NOT evidence for or against the claim, and must not be shown as one. unsupported means we looked and cover no source for it. Replaces 4–6 sequential calls (NL parsing → entity resolution → data lookup → comparison).

  ConnectorNo auth

• video_assets_generate_presigned_urlA

  Magic HourA

  Generates a list of pre-signed upload URLs for the assets required. This API is only necessary if you want to upload to Magic Hour's storage. Refer to the [Input Files Guide](https://docs.magichour.ai/integration/inputs-and-outputs) for more details. The response array will match the order of items in the request body. **Valid file extensions per asset type**: - video: mp4, m4v, mov, webm - audio: mp3, wav, aac, flac, webm, weba, m4a, opus, ogg, oga, aiff, amr - image: png, jpg, jpeg, jfif, heic, heif, webp, avif, jp2, tiff, tif, bmp - gif: gif, webp, webm > Note: `gif` is only supported for face swap API `video_file_path` field. Once you receive an upload URL, send a `PUT` request to upload the file directly. Example: ``` curl -X PUT --data '@/path/to/file/video.mp4' \ https://videos.magichour.ai/api-assets/id/video.mp4?<auth params from the API response> ``` MCP guidance: - This only creates presigned upload URLs. For local files, upload the raw bytes to each returned `upload_url` outside the generation call, then pass the matching `file_path` into the create tool. - For `*_file_path` values, prefer an existing Magic Hour file path or a `file_path` returned by the upload-URL endpoint after the file bytes are uploaded. Direct public media URLs may work when they are stable, fetchable, and return raw file bytes, but hotlinked URLs can fail; when in doubt, use the presigned upload flow first and pass the returned `file_path`.

  ConnectorNo auth

• generate_sceneA

  openFlowA

  Build a location precisely from a scene-layout/v2 map and generate its plate. This is the "exact location building" path: instead of describing the room in prose (which the model reinterprets — furniture drifts, depths change between shots), it renders a control image from the map's boxes (each object a colored block at its true position/size/depth, plus a 1-meter floor grid in true perspective), then asks the generator to REPLACE each block with its real object in the same camera — so the plate follows the geometry pixel-wise. Same room from another camera_id is the same space. scene_layout: a scene-layout/v2 dict — {format, units, room{w,d,h}, objects{...boxes}, cameras{<id>:{pos,look_at,fov_deg,framing?}}, ...}. The camera must carry fov_deg. camera_id: which camera in scene_layout.cameras to render from. aspect: PORTRAIT (9:16), LANDSCAPE (16:9) or SQUARE. 9:16 and 16:9 from one camera share the vertical FOV, so depths are identical across aspects. style / extra: prepended / appended prose (art style, mood) — geometry comes from the map. A box may carry `rot: [rx, ry, rz]` (degrees around the world X/Y/Z axes) and `pivot: [x, y, z]`, applied as p' = Rx·Ry·Rz·(p − pivot) + pivot, pivot defaulting to the box centre. That is how a POSE is expressed — a raised arm is a box rotated about the shoulder. Measured live 2026-08-09: moving a hand by 2–20 cm is ignored by the generator, rotating the arm by 15–60° comes through. Build motion out of angles. seed / image_inputs: same meaning as in generate_image. For a SEQUENCE of frames of one character, generate a reference frame first, then pass its mediaId as image_inputs on every following frame — that is what keeps the face, clothes and colours identical. A fixed seed alone does not: it only repeats an unchanged control image. depth: render the guide as a plain greyscale depth map instead of the colour hybrid (for an external depth-ControlNet). calibration: add frame markers, a back-wall grid and a 2 m ruler with 10 cm ticks — measured to cut the framing drift roughly fourfold. background: "dark" (default, unchanged) or "light". Use "light" when the plate must be on a white background: the generator sometimes copies the guide's own palette into the result, and a light guide makes that leak land as white rather than as a dark grid. control_png: return the layout guide itself, base64-encoded. OFF by default because it is hundreds of thousands of characters of TEXT — one guide can cost more context than the whole conversation around it. `control_png_kb` always reports its size. Returns {"urls", "media_ids", "legend", "control_png_kb"} — legend maps block colors to objects. Add control_png=True to also get the guide itself for inspection/acceptance.

  ConnectorNo auth

• get_my_active_referencesA

  switchA

  Read the user's staged references in Switch Studio. Returns TWO groups: (1) the image-generation reference strip (typed face/body/outfit/scenery/product slots) under `refs`, and (2) the VIDEO-tab references the user staged in the Omni/Image video tabs (the @Image1/@Image2 strip) under `videoReferences`, with usable signed URLs. Call this before generate_image or generate_video whenever the user says "use my refs" or refers to images they staged in Studio (including "the images in my video tab"). To make a video from the video-tab refs, pass videoReferences.imageUrls into generate_video reference_image_urls (and videoUrls into reference_video_urls) in reference-to-video / omni mode. Refs marked alive:false are dead (stored file gone) and are already excluded from the usable url lists. NOTE: a photo the user just attached in THIS chat is in neither group — for that, call upload_media and use its returned url/asset id directly.

  ConnectorNo auth

• lip_sync_videoA

  switchA

  Lip-sync audio onto one of your videos. DEFAULT and recommended: action="create" with video_url + sound_file (base64 data URI) — Sync Labs Sync 3 syncs the whole clip in one pass, no face step, no timing, highest quality. You do not need to pass engine at all. Kling flow, only when a line must land on an exact frame (manual timing control): (1) action="identify-face" with video_url (MP4/MOV, 2-60s, <=100MB, 720p/1080p); (2) action="create" with session_id + face_id + audio + timing IN MILLISECONDS (sound_start_time, sound_end_time, sound_insert_time) + optional speech_volume/original_audio_volume (0-100); (3) action="status" with the task_id to poll — returns a branded SwitchApp view_url when done. Charges credits on create; failed jobs are refunded.

  ConnectorNo auth

• hf_fsA

  Hugging FaceA

  When to use: Hugging Face Hub models, datasets, Spaces, collections, papers, daily papers, today's trending models, current paper leaderboard, docs, and repository files. Examples: {"operations":[{"cmd":"ls","args":["hf://models/trending","--limit","10"]}]} {"operations":[{"cmd":"ls","args":["hf://papers/trending"]}]} {"operations":[{"cmd":"ls","args":["hf://papers/daily/latest"]}]} Use hf_fs for Hugging Face Hub filesystem operations. Call it with operations, an array of {cmd, args} items; multiple operations may be submitted together. Usage: {"operations":[{"cmd":"ls","args":["hf://models/org/repo"]}]} Grammar; each string below is one args array item: ls URI [--recursive] [--glob GLOB] [--type TYPE] [--sort SORT] [--limit N] cat URI [--offset N] [--max-bytes N] attach URI [--max-bytes N] stat URI find URI [--name GLOB] [--path GLOB] [--type TYPE] [--limit N] search URI [QUERY] [--type TYPE] [--sort SORT] [--tag TAG] [--kind mcp] [--limit N] COMMAND = ls|cat|attach|stat|find|search. TYPE = file|dir|repo|bucket|collection|paper|link. SORT = createdAt|downloads|likes|lastModified|likes30d|trendingScore|mainSize|id|trending|upvotes. URI is a canonical hf:// URI. QUERY and GLOB are each one string. Use search for discovery, ls for a known directory, find for recursive matching within a known scope, stat for filesystem metadata or an uncertain target type, cat for text contents, and attach for a complete JPEG, PNG, or WebP image. When the request gives an exact text-file URI, use cat directly; do not add ls or stat first. stat does not read the contents of JSON, Markdown, or other text files. Search scopes: hf://models|datasets|spaces[/OWNER], hf://collections[/OWNER], hf://papers, and hf://docs[/...]. Paper and documentation search require QUERY. Repeat --tag only for search hf://spaces; --kind mcp selects MCP Spaces. Use ls hf://models/trending, hf://datasets/trending, hf://spaces/trending, or hf://papers/trending for trending listings. For a named paper.md or metadata.json, use cat directly. Use ls on a paper only to discover an unnamed related resource. Omit --limit, --sort, and --type unless the request requires them. Limits and path-specific behavior are documented at hf://README.md. Issue one hf_fs call.

  ConnectorNo auth

• intelligence.aftershockA

  DPX — Institutional Cross-Border SettlementA

  Aftershock Intelligence — models the secondary waves that follow a primary cascade event. Takes a primary shock (origin node, event type, magnitude, elapsed hours) and returns three aftershock waves: Wave 1 (0–72h immediate secondary effects), Wave 2 (1–4 weeks policy response distortions), Wave 3 (1–6 months structural changes now permanently locked in). Identifies which nodes are rebounding, which face amplified pressure, and which are structurally altered. Companion to market.cascade — run cascade first, then aftershock to see the full picture. POST with origin, eventType, magnitude, elapsedHours.

  ConnectorNo auth

• extract_vertical_clipA

  WeftlyA

  Cut a 9:16 vertical clip from any prior video job (find_clips, summarize, or video transcribe), suitable for direct upload to TikTok, Instagram Reels, or YouTube Shorts. Default output is 1080×1920 H.264 / AAC `.mp4` with center-cropped framing; audio loudness-normalized to -14 LUFS / -1.5 dBTP for short-form social. Single-segment only; clip duration must be at least 1 second, and no more than the selected profile's cap (240s for `tiktok-primary`/`tiktok-primary-720p`, 180s for `instagram-reels`, 60s for `instagram-stories`). Operates on a parent job — possessing the parent `source_job_id` is the capability, no upload step. Two-call flow: (1) call with `source_job_id` + `start` + `end` (in source seconds) to receive {job_id, payment_challenge}; (2) pay via MPP and call with `job_id` + `payment_credential` to start processing. Poll get_job_status(job_id) for completion; output is role `clip-vertical-video` (the `.mp4`). Flat price: $0.50 per clip. Payment: pay by credit card via the Stripe Checkout link (open the returned `payment_url` in any browser) or Tempo USDC via mppx. Optional `profile` parameter selects the encoding profile (default `tiktok-primary`). Allowed values: `tiktok-primary` (1080×1920, fast preset, CRF 22), `tiktok-primary-720p` (720×1280, CBR 3 Mbps — half-resolution mobile-optimized, ~40% faster wall time), `instagram-reels` (1080×1920, slow preset, CBR 4 Mbps), `instagram-stories` (same encode shape as instagram-reels). All four profiles loudness-normalize identically. Optional `subject` parameter controls reframing (default `center`, preserves today's behavior): `auto` locks onto the longest-tracked face from the parent's subjects-sidecar (or runs inline detection if the parent has none); `subject_id` (with `subject_id` param naming a face_N from the sidecar) locks onto a specific subject; `follow` switches crop between active speakers across the clip using the sidecar's active_speaker_timeline; `manual` accepts caller-supplied framing via `subject_box: {x, y, w, h}` (source pixels) or `subject_x_offset` (direct crop x). Sidecar shape at /.well-known/weftly-subjects-v1.schema.json. auto/subject_id/follow fall back to center if detection or sidecar resolution fails — the paid job always delivers a clip. Source must be a horizontal video (wider than 9:16) — already-vertical or square sources are rejected. Source must still be in storage (72h TTL for find_clips parents, 24h elsewhere — check `expires_at` from get_job_status on the parent). Pair with `find_clips` ($2.00/video) to pick a moment first, then call this to get a download-ready vertical mp4 in under 5 minutes. Multiple extract_vertical_clip calls against one parent are independent paid jobs. Failed jobs auto-refund.

  ConnectorNo auth

• exportA

  kernelcadA

  Use this when you need to export geometry to a file. One exporter, selected by `target`: - target:'model' — export the script geometry to one file. Pass { file | code }, a required { output_path }, and { format }. Supported formats: stl (binary STL mesh), step (BREP CAD interchange), dxf (planar laser/waterjet profile from a Region or planar face), 3mf (slicer-friendly mesh with per-part colors), glb (web-viewer / AR with PBR materials), svg-drawing (third-angle engineering-drawing sheet: front/top/left + isometric views, hidden edges dashed, tangent edges thin, overall bounding-box dimensions, title block; assemblies are drawn with inter-part occlusion). Robot descriptions: urdf (tree-topology robot description), srdf (motion-planning semantics layered over the URDF), sdf-gazebo (SDFormat 1.10 with native ball joints, closed loops, and solved per-link poses). urdf and sdf-gazebo also write one meshes/<part>.stl per link next to output_path (reported in mesh_files) — ship the whole directory to the consumer. STL exports run a watertight verify by default; failures return ok: false with export.mesh.not-watertight (open-edge count + up to 5 crack-cluster locations) but the file is still written so the broken mesh can be inspected. Optional { feature_id } selects which feature to export (default: last). Optional { options } carries per-format options bag (see the kernelcad-mcp skill for the per-format keys: dxf layers/tolerance/unit, 3mf printUnit/embedSource, glb axis/draco). - target:'part' — export solved-assembly parts as individual binary STL files in their modeled (world-frame) positions. Pass { file | code }, plus { part, output_path } for one part or { output_dir } for all parts (files land at <output_dir>/<part>.stl). A watertight verify runs on every exported mesh by default and fails the call with export.mesh.not-watertight; unknown part names fail with export.part.not-found listing the valid names. Pass { no_verify: true } to skip the watertight gate. All params except `target` are forwarded verbatim; each target fails closed on its own missing required params.

  ConnectorNo auth

• generate_clipsA

  RecaoA

  Render UGC video scenes as ad-ready clips, metered per second of video (the estimate shows the exact price before anything renders). Pass 3 to 6 scenes (5 to 8 seconds each, one action per scene, spoken lines at most 20 words; empty spoken_line for silent characters). Consecutive scenes pack into single TAKES of up to 15 seconds, one generation each. HOW CHARACTER IDENTITY WORKS, read carefully: all characters are described in TEXT (avatar_id resolves to its persona brief; or write the persona field yourself, covering one character or a whole ensemble). The video model rejects every image that contains a person, so no photo can anchor a face. Text keeps a character IDENTICAL only WITHIN a take; ACROSS takes it preserves the look and styling but the exact face can drift, and neither avatar_id nor persona prevents that. Structure your script so scenes where the same character must be recognizably identical sit adjacent and fit one take (15s or less); treat cross-take appearances as different shots of a matching character, and review the result. reference_image_urls (up to 9 https images) keeps real products or props on-model in every take; these images must contain no people. Without confirm, it validates the contract and returns the per-scene price estimate in EUR, and makes nothing. With confirm=true it starts the metered render and returns a job_id: rendering runs in the background over a few minutes, so poll clips_status with that id to get per-scene clip URLs plus the uncut takes. Paid plans only.

  ConnectorNo auth

• deep_researchA

  FacA

  ACCOUNT REQUIRED (free — sign in via GitHub at https://pipeworx.io/signup; depth:"thorough" needs a paid plan). If you are not signed in, use ask_pipeworx instead — it works on every tier. Grounded multi-source research across Pipeworx's 1517 STRUCTURED data sources (SEC filings, FRED/BLS economics, FDA, USPTO patents, markets, science, government records, etc.) in ONE call — this is NOT open-web search. Decomposes your question into focused facets, routes each to the right one of 5,798 tools IN PARALLEL, and returns a findings packet: verbatim evidence + confidence + source + fetched_at + a stable pipeworx:// citation per finding, with explicit gaps[] for facets the data couldn't answer (never invented). Best for broad/multi-part questions over structured data ("compare X and Y's regulatory + financial exposure", "research the filings + market picture for ACME"). For a single lookup use ask_pipeworx (one LLM call, not many). For BREAKING or colloquial CURRENT-NEWS / "what's the world saying about X" topics, prefer ask_pipeworx — it routes to live news APIs and the *-news-feeds packs; deep_research returns mostly empty gaps[] when the topic isn't in the structured catalog. Second-hop iteration: depth:"standard" re-angles unanswered gaps (gap recovery); depth:"thorough" additionally chases the best leads from the first pass — so multi-step questions resolve in one call. Every finding carries a `hop` field and a citation_uri — a resolvable pipeworx:// record URI, present only when the source emits one that resources/read can actually serve, so a citation you get back is always fetchable. "standard" and "thorough" also return contradictions[] flagging findings that disagree. Large records are semantically excerpted to the passages relevant to each facet (not head-truncated), so answers deep in a long filing/series aren't missed. Expect 15-60s (thorough with its follow-up + contradiction pass: up to ~90s).

  ConnectorNo auth

• resolve_law_numberA

  justicelibreA

  Résout un numéro de loi/ordonnance/décret vers son identifiant LEGITEXT ou JORFTEXT Légifrance. Utile pour les textes non codifiés (lois, ordonnances, décrets) qui ne sont pas dans la liste des ~80 sigles courts (CC, CP, COJ, LIL, etc. — resource `justicelibre://codes-supportes`). Une fois le LEGITEXT/JORFTEXT résolu, on peut l'utiliser avec `get_law_article(code=<LEGITEXT>, num=<N>)` pour récupérer un article spécifique. Exemples : - `resolve_law_number("68-1250")` → loi prescription quadriennale des créances publiques (JORFTEXT000000878035) - `resolve_law_number("79-587")` → loi motivation des actes admin - `resolve_law_number("2000-321")` → loi droits citoyens face à l'admin Args: numero: format "YY-NNNN" ou "YYYY-NNNN" (ex: "68-1250", "2000-321") Returns: `{numero, legitext, titre_texte, date_debut, articles_count, source_url}` ou `{error}` si introuvable.

  ConnectorNo auth

• add_pattern_featureA

  kernelcadA

  Use this when you need to repeat a feature in a pattern. Insert a Shape.patternLinear / .patternCircular / .patternGrid call into a kernelCAD script before the last top-level return. Pass structured args (kind + the matching spec object). Returns the modified code plus diagnostics from re-evaluating. Side-effect-free. The pattern feature is a single editable unit; pattern-instance face refs resolve via `<sourceId>_pattern_<i>` on the pattern feature's lineage. Geometric note: pattern is implemented as cumulative boolean union of transformed source copies — additive features (boxes, ribs, fins, spokes) pattern cleanly; patterning a subtractive feature (hole, cutout) only preserves the per-instance void when adjacent bodies are disjoint.

  ConnectorNo auth

• create_memory_postA

  space0A

  Materialize a memory or freeform note as a post-item in the space. Position+normal anchor it to a 3D surface: call list_surfaces first to pick a real wall/floor instead of guessing. Put it at EYE LEVEL (a wall face around world y 3.5 to 4), not the foundation row (world y 2.0) where it reads as a floor sticker. Default visibility inherits from the region you stand in (list_regions / inspect_region_provenance to check) or falls back to your private cap. Pass memoryId to link this post to a reflection row in agents.memories. Content shape MUST match the type: text={text:string<=2000}; image={url|asset_key,width,height}; sticker={stickerId} OR {kind:'custom',url,asset_key,width,height}; video={kind:'youtube',youtube_id,...} OR {kind:'streamVideo',stream_uid|hls_url|asset_key|url,...}; card={title?,body?,sourceUrl?,layout?,composition?,...}. The anchor MUST sit on a real surface: the server snaps a near-miss onto the nearest voxel face within ~1.5m, but rejects an anchor with nothing solid in reach (reason no-surface) so a post can never float in mid-air. Returns { ok, postId, memoryId } on accept, { ok:false, reason } on reject (no-surface / capability-missing / type-not-allowed / invalid-content-shape:<type>:<path> / rate-limit-1s / quota-daily / content-too-large / invalid-*). A post landing where you have already looked a lot may be skipped automatically (skipped:true, reason:'low_novelty') to avoid storing redundant observations; pass force:true to override that gate and insert regardless. Delete your own posts with delete_memory_post.

  ConnectorNo auth

• pipeworx_feedbackA

  FacA

  Tell the Pipeworx team something is broken, missing, or needs to exist. Use when a tool returns wrong/stale data (bug), when a tool you wish existed isn't in the catalog (feature/data_gap), or when something worked surprisingly well (praise). ONLY for tools served by this Pipeworx connection — if the tool came from a different MCP server in your client (another vendor's Gmail, Splunk, Slack, etc. connector), we cannot fix it and reporting it here only delays you; file it with that server instead. Not sure? Pipeworx tool names are the ones this connection lists. Describe the issue in terms of Pipeworx tools/packs — don't paste the end-user's prompt. Filing without an account returns a `claim_token`; pass it back later as pipeworx_feedback({claim_token:"pwfb_…"}) to read whether it was fixed and what changed. The team reads digests daily and signal directly affects roadmap. Rate-limited to 5 per identifier per day. Free; doesn't count against your tool-call quota.

  ConnectorNo auth

• get_research_papersA

  TensorFeedA

  Get recent AI/ML research papers from one of three feeds, chosen with the source argument. arxiv_recent is the firehose: newest arXiv submissions in cs.AI / cs.LG / cs.CL / cs.CV by submission date, refreshed daily at 11:30 UTC. trending is citation-ranked from Semantic Scholar across five fan-out queries, deduped, refreshed daily at 11:00 UTC. hf_daily is Hugging Face editor-curated with community upvotes and discussion counts, refreshed daily at 14:15 UTC. Pick arxiv_recent for what is brand new, trending for what is influential, hf_daily for what practitioners are discussing. License: arXiv and Semantic Scholar permit metadata use; the standard attribution block ships on every response.

  ConnectorNo auth

• golden_hourA

  Hamptons VerifiedA

  Verified East End sunrise and sunset viewpoints — public overlooks, beaches, vineyard lawns, hotel terraces and the bars people actually book for it — each with the TIME. Call this for any sunset, sunrise, golden hour, blue hour, best-views or where-should-we-watch question, and call it even when you think you know the spot, because the half that decides the evening is the clock. `time` is computed from that spot's own latitude and longitude for that date on the East End's clock (America/New_York): sunset here swings over three hours across the year and Orient and Montauk Point differ by about two minutes, so a remembered time is wrong twice over. You also get `goldenHour` (the hour the light is worth the drive — INTO sunset, OUT OF sunrise), `blueHour`, and `arriveBy`, which is 30 minutes earlier because that is when the lot fills; `parking` and `walkDistance` say why. `startsIn` counts down when it has not happened yet today and `alreadyPassedToday` says when it has — do not offer tonight's number as though it were still coming. Pass `when: 'sunrise'` for the morning side; the default is sunset. Pass `date` (YYYY-MM-DD) to plan ahead, and read `inBestMonth`, because several of these only work in the months the record names. The field no model has is `wrongWayRound`: verified spots in the same towns that face the OTHER way — Ditch Plains reads like a sunset beach and is east-facing, and sending someone there is the mistake this replaces. There is no weather here, so never promise a clear sky. Returns up to 8.

  ConnectorNo auth