3 019 media skills from 496 authors. They create and process images, video and sound. Half of them fit into 1 966 tokens or less — that is what one costs your context window when the agent loads it. 768 ship runnable scripts rather than instructions alone. 30 of them cannot work without an MCP server, most often rube. We also found 321 copies of these same skills sitting in other people's repositories — counted once here, not 321 times.
3 019 unique 496 authors 1 925 updated this month 160 from vendors
> Use when the user asks to "analyze video", "watch this video", "what happens in this video", "describe this clip", "review this footage", "classify these videos", "compare videos", "analyze this image", "what's in this screenshot", or when the user provides a video/image "multimodal analysis", "motion analysis", "video reference", "video breakdown", "batch classify", or any task requiring understanding of video content that Claude cannot do natively.
> Scaffold and author content for the microsoft/agent-academy GitHub repo. Use this skill whenever the user wants to create a new mission, lab, or module for Agent Academy — whether that's a course mission (Recruit, Operative, Commander) or a standalone Special Ops mission. Trigger whenever the user says "write a mission", "scaffold a lab", "create a Special Ops", "new Agent Academy module", or describes an agent-building concept they want to turn into a hands-on lesson. Also trigger when the user says "add a module to agent academy", "I want to contribute a mission", duplicate detection, mission naming, markdown scaffolding, step-by-step content drafting, cross-linking, tag/category suggestions, difficulty rating, and badge image generation for Special Ops missions.
Extracts YouTube transcripts and generates professional Flutter showcase blog posts in Markdown format. Use when you need to create a technical marketing case study from a video success story.
Validates that all TypeSpec route operations in the AI Foundry data-plane (Foundry) domain have documentation comments and @summary decorators with correct voice, replaces documentation-required suppressions with generated documentation, and uses description override extensions when TypeSpec would concatenate noisy descriptions. USE FOR: reviewing or authoring routes.tsp and routes.generated.tsp files under specification/ai-foundry/data-plane/Foundry/src/, including nested OpenAI routes under src/openai/. DO NOT USE FOR: files outside the Foundry data-plane area, model-only .tsp files, or SDK client customization files (client.tsp).
Use when the user wants to search, query, extract, transcribe, describe, quote, filter, or aggregate across documents — PDFs, scanned forms / images (`.jpg` `.png` `.tiff`), Office (`.docx` `.pptx`), text (`.html` `.txt`), audio (`.mp3` `.wav` `.m4a`), or video (`.mp4` `.mov`). Prefer this over native Read / Grep for multi-file or non-PDF corpora. Not for: editing files, web browsing, single-file plain-text lookups, fine-tuning.
Provides digital forensics and signal analysis techniques for CTF challenges. Use when analyzing disk images, memory dumps, event logs, network captures, cryptocurrency transactions, steganography, PDF analysis, Windows registry, Volatility, PCAP, Docker images, coredumps, side-channel power traces, DTMF audio spectrograms, packet timing analysis, CD audio disc images, or recovering deleted files and credentials.
Provides miscellaneous CTF challenge techniques for problems that do not cleanly fit the main categories. Use for encoding puzzles, pyjails, bash jails, RF/SDR, DNS oddities, unicode tricks, esoteric languages, QR or audio puzzles, constraint solving, game theory, unusual sandbox escapes, and hybrid logic puzzles. Prefer a more specific skill first when the challenge is mainly web, pwn, reverse, forensics, malware, OSINT, or crypto. Treat this as the fallback skill for genuine cross-category or edge-case challenges, not the default starting point.
Provides open source intelligence techniques for CTF challenges. Use when gathering information from public sources, social media, geolocation, DNS records, username enumeration, reverse image search, Google dorking, Wayback Machine, Tor relays, FEC filings, or identifying unknown data like hashes and coordinates.
Build call automation workflows with Azure Communication Services Call Automation Java SDK. Use when implementing IVR systems, call routing, call recording, DTMF recognition, text-to-speech, or AI-powered call flows.
| Azure AI Content Understanding SDK for Python. Use for multimodal content extraction from documents, images, audio, and video.
Build real-time voice AI applications using Azure AI Voice Live SDK (azure-ai-voicelive). Use this skill when creating Python applications that need real-time bidirectional audio communication with Azure AI, including voice assistants, voice-enabled chatbots, real-time speech-to-speech translation, voice-driven avatars, or any WebSocket-based audio streaming with AI models. Supports Server VAD (Voice Activity Detection), turn-based conversation, function calling, MCP tools, avatar integration, and transcription.
|- Azure Speech to Text REST API for short audio (Python). Use for simple speech recognition of audio files up to 60 seconds without the Speech SDK.
Analyze text and images for harmful content using Azure AI Content Safety (@azure-rest/ai-content-safety). Use when moderating user-generated content, detecting hate speech, violence, sexual content, or self-harm, or managing custom blocklists.
Bluetooth Classic (BR/EDR) attack methodology — device discovery, service enumeration via SDP, LMP/L2CAP layer attacks, legacy PIN cracking (BlueBorne / KNOB), Bluetooth file-transfer abuse (BlueSnarfing legacy), unauthenticated profile abuse (HSP, HFP, OPP), and modern relevance against older industrial / automotive / accessory targets. Use when in-scope devices use Bluetooth Classic (Bluetooth ≤ 4.0 BR/EDR) — common in legacy car kits, industrial sensors, older medical devices, and audio accessories.
Stage 1 of Clinical ASR Flywheel. Use when bootstrapping a cycle: NVCF+MW disclosure, NVIDIA_API_KEY check, deps install, TTS+ASR smoke test.
> Use this skill when the user is doing hands-on DOCA DPA host-side work on a BlueField — creating the `doca_dpa` Core context, loading a DPACC-compiled DPA app image (`doca_dpa_app`), creating DPA threads, launching kernels via `doca_dpa_kernel_launch_update_*`, draining `doca_dpa_completion`, running `doca_dpa_cap_*` discovery, choosing between the DPA comm component (inter-DPA messaging) and the DPA verbs component (in-kernel RDMA), or debugging `DOCA_ERROR_*` from `doca_dpa_*`. Trigger even without "DOCA DPA" or "Data-Path hangs, no completion", "DOCA_ERROR_DRIVER on launch", "DOCA/DPACC version skew", or "does this BlueField expose a DPA". Route elsewhere for DPA-side kernel programming itself, DPACC compiler internals, host↔DPU messaging (doca-comch), host-side RDMA (doca-rdma), and GPU-initiated networking (doca-gpunetio).
> Use this skill when the user is doing hands-on DOCA Rivermax work on a BlueField DPU or ConnectX host — standing up `doca_rmax_in_stream` (receive) sessions for timing-precise media-over-IP (SMPTE ST 2110 video/audio, market data, scientific feeds), confirming the Rivermax SDK + license precondition before any DOCA-side code, running `doca_rmax_get_*_supported` capability queries, pairing with `doca-eth` queues and `doca-flow` steering, or debugging `DOCA_ERROR_*` from a Rivermax call. Trigger even when the user does not explicitly mention "DOCA Rivermax" or "rmax" — implicit phrasings include "ST 2110 receive isn't getting frames", "sub-microsecond jitter on BlueField", "NOT_SUPPORTED from doca_rmax_init", "no recv events after stream start", or "license check failing on a media receiver". Refuse and route elsewhere for installing the Rivermax SDK or its license, programming the underlying queue (`doca-eth`), steering rules (`doca-flow`), or best-effort packet I/O — those belong to other skills.
Render a single DRR fluoroscopy frame from a CT cache or synthetic phantom. Use when asked to render DRR, generate a fluoro image, or smoke-test the Slang renderer.
Use to flash a promoted BSP image to a Jetson DUT in RCM mode via flash.sh or l4t_initrd_flash.sh. Do NOT use for BSP customization, image promotion, or carrier derivation.
>- Build a per-target knowledge-base markdown next to the active profile by walking the BSP root and source tree. Use after init-image / init-source; not for editing profile fields.
>- bsp_sources, Crosstool-NG toolchain. Use after jetson-init-image; not for fetching inputs.
>- Extract Jetson Linux + sample-rootfs tarballs and run apply_binaries.sh for the active target, then record bsp_image in the profile. Use after jetson-init-target; not for source-tree setup.
Pick Jetson-compatible containers, vLLM runtime images, and Jetson AI Lab PyPI indexes; maps Orin SM 8.7 vs Thor SM 11.0 and JetPack-specific package choices.
>- Use to promote overlay files and built artifacts into the staged image.
>- Use after jetson-flash-image to run static BSP checks, on-target smoke/regression tests on a flashed DUT, or both. Not for build
Routes NVIDIA Nemotron Speech (Riva) NIM tasks — deploys, runs, and tests ASR, TTS, and NMT NIMs on build.nvidia.com or self-hosted.
>- Use when the user wants to orchestrate defect image generation with NVIDIA Cosmos AnomalyGen (Cosmos-Predict2-derived) on OSMO for PCBA, metal surface, and glass inspection. The Day 0 path handles cold-start with USD-to-ROI, image-edit augmentation, and AnomalyGen to create initial PCBA datasets. The Day 1 path performs inference and labeling on real images. This skill helps with first-time asset setup, creation of finetuning checkpoints, and configuring deployment.
>- Use when running people attribute search (PAS) image augmentation and attribute search, PAS, person augmentation, attribute search, person re-identification, clothing augmentation, person crop augmentation.
>- flow selection, preflight, submit-time interpolation, monitoring, and output labeling, VDA demo, OSMO workflow, pseudo labeling.
Performs deep Root Cause Analysis (RCA) on NVIDIA TAO Visual ChangeNet classification experiments with image-evidence-driven investigation. Use when analyzing ChangeNet model failures, investigating poor recall / FAR / PASS-NO_PASS metrics, auditing visual inspection pipeline quality, or running an RCA report for an AOI defect-detection model. Trigger phrases include "RCA on my ChangeNet model", "why is my AOI model failing", "audit ChangeNet predictions", "investigate FAR regressions", "root cause analysis on visual-changenet".
Apply interface craft when building or reviewing UI - motion, easing, timing, springs, component feel, and visual foundations. Use when building a component, animation, transition, hover or press state, modal, drawer, toast, or when polishing an interface so it feels right. Says "make this feel better", "add an animation", "polish the UI", "review this component".
Generate or edit images using AI models (FLUX, Nano Banana 2). Use for general-purpose image generation including photos, illustrations, artwork, visual assets, concept art, and any image that is not a technical diagram or schematic. For flowcharts, circuits, pathways, and technical diagrams, use the scientific-schematics skill instead.
Lightweight WSI tile extraction and preprocessing. Use for basic slide processing tissue detection, tile extraction, stain normalization for H&E images. Best for simple pipelines, dataset preparation, quick tile-based analysis. For advanced spatial proteomics, multiplexed imaging, or deep learning pipelines use pathml.
Use when the user asks to generate or edit images via the OpenAI Image API (for example: generate image, edit/inpaint/mask, background removal or replacement, transparent background, product shots, concept art, covers, or batch variants); run the bundled CLI (`scripts/image_gen.py`) and require `OPENAI_API_KEY` for live calls.
Microscopy data management platform. Access images via Python, retrieve datasets, analyze pixels, manage ROIs/annotations, batch processing, for high-content screening and microscopy workflows.
Computational pathology toolkit for analyzing whole-slide images (WSI) and multiparametric imaging data. Use this skill when working with histopathology slides, H&E stained images, multiplex immunofluorescence (CODEX, Vectra), spatial proteomics, nucleus detection/segmentation, tissue graph construction, or training ML models on pathology data. Supports 160+ slide formats including Aperio SVS, NDPI, DICOM, OME-TIFF for digital pathology workflows.
Python library for working with DICOM (Digital Imaging and Communications in Medicine) files. Use this skill when reading, writing, or modifying medical imaging data in DICOM format, extracting pixel data from medical images (CT, MRI, X-ray, ultrasound), anonymizing DICOM files, working with DICOM metadata and tags, converting DICOM images to other formats, handling compressed DICOM data, or processing medical imaging datasets. Applies to tasks involving medical image analysis, PACS systems, radiology workflows, and healthcare imaging applications.
Use when the user asks to create or edit videos end-to-end (script→video, auto-cut/jumpcut, captions/subtitles, polishing for Shorts/Reels/TikTok). Current implemented backend: local FFmpeg (probe/render/jumpcut/burn-subtitles/polish). Planned/optional backends: Remotion (motion graphics templates), VectCutAPI (CapCut/剪映 timeline editing), and video-audio-mcp (MCP tool wrapper) when available. Produces a finished video artifact (MP4 by default) from assets + copy + a design/storyboard plan.
Use when the user asks to "codify our brand voice", "define naming rules for our products and tiers", or "write the tone-of-voice guide with banned phrases"; produces the brand-level voice canon (register, tone spectrum, banned-phrase list, few-shot examples drawn only from the brand''s own published material) and the naming tax (product / feature / tier naming rules plus approved and banned terms) that seeds the narrative-registry canon and that every channel''s voice adaptation points up to. Not for per-platform voice adaptation — use channel-registry''s voice-dossier; not for finished copy or blog posts — use content-writer; not for the message hierarchy itself — use message-system-architect; not for claim adjudication — use offer-claims-registry. 品牌语气/词汇表/命名税/禁用词/品牌语言规范
Use when the user asks to "build a story bank", "collect our origin and customer stories", or "assemble reusable proof stories for the message"; assembles reusable narrative units — origin, founder, customer, transformation, and proof stories — each tagged to a claims-ledger ID and a message-house pillar, with every proof labeled Measured / User-provided / [needs source]. Not for authoring the message house or pillars — use message-system-architect; not for brand voice or naming rules — use brand-language-codifier; not for finished long-form prose — use content-writer; not for adjudicating whether a proof is true — use offer-claims-registry. 品牌故事库/起源客户转化/证据故事单元
Use when the user asks to "make everyone tell the same story", "write our elevator pitch ladder", or "build a spokesperson Q&A and approved boilerplate pack"; derives from the narrative-registry canon, message house, and brand voice a repeatable enablement kit — an elevator ladder (10-second / 30-second / 2-minute), a spokesperson Q&A (tough questions with on-canon answers), an approved boilerplate/bio pack (25 / 50 / 100-word), and a do/don''t language sheet — so sales, support, founders, and partners repeat one consistent story. Not for the launch-day runbook — use launch-day-conductor; not for finished per-channel copy — use content-writer or each discipline creative builder; not for launch-window battle cards — use sales-enablement-kit; this kit never adjudicates a claim. 叙事赋能包/电梯梯度/发言人问答/审定样板/该说不该说
Use when the user asks to "map what our buyers believe", "capture the objections we keep hearing", or "find the switching forces that move the beachhead"; produces a belief map of the beachhead — held beliefs and mental models, the recurring objections and their reframes, and the JTBD four forces (push of the problem, pull of the new, anxiety of switching, habit of the present) — each item sourced from interviews or win-loss notes (User-provided) and labeled Measured / User-provided / Estimated, with any unverified quote or comparative claim marked "[needs source]" and routed to the claims candidates, never adjudicated here. Not for demographic or persona profiling — use audience-mapper; not for the positioning canvas — use positioning-truth-tracer. 受众信念/异议地图/切换四力/流失语言
Use when the user asks to "check our positioning against what we can actually ship", "trace which differentiators we can defend", or "reconcile the positioning canvas with the claims ledger"; reconciles the reused positioning canvas against the shippable stage and the claims ledger to produce a differentiation truth set — every differentiating claim verifiable or marked ''[needs source]'' — that TALE-T1 is judged against. Not for building the canvas — use positioning-mapper; not for adjudicating claims — use offer-claims-registry; not for authoring the message house — use message-system-architect. 定位真相/差异化校准/可交付现实/主张核对
Use when the user asks to register/query a social channel, record channel state, cadence, governance, voice adaptation, UGC permission, or advocacy facts; curates them through the append-only channels event stream and derived views. Not for ECHO scoring — use social-quality-auditor; not for channel selection — use channel-portfolio-planner. 渠道台账/账号档案/UGC授权记录
Use when the user asks to record/query the brand narrative canon, tagline, message hierarchy, voice/naming rules, or a canon re-version; curates complete versioned canon events through the append-only narrative stream and derived views. Not for TALE scoring — use narrative-quality-auditor; not for authoring the system — use message-system-architect. 品牌叙事台账/canon 记录/语气与命名规范
Use when the user asks to "script this short video", "write a TikTok / Reels / Shorts script", "给这条抖音或视频号视频写脚本", or "fix the hook — viewers drop off in the first seconds"; produces timestamped beat-sheet scripts on the retention-gate model (0-2s hook / 2-5s confirmation / 5-15s payoff / loop-or-CTA) with per-scene script lines, on-screen text for muted viewing, asset keywords, and 9:16 format-param rows per platform norm card for TikTok, Reels, Shorts, 抖音, and 视频号 — plus 2-3 hook options from named hook families and an AI-content disclosure line by default on realistic synthetic media. Spec-only: rendering, TTS, and publishing stay with the user''s own tools — no pipelines, no upload automation. Not for creator video briefs — use brief-generator; long-form video SEO belongs to content-writer. 短视频脚本/抖音分镜/开头钩子/竖屏9:16
Use when the user asks to "codify how our founder sounds", "build a founder voice dossier", or "set our content pillars"; runs an 80%-extraction interview over the user''s OWN posts, emails, and decks (never competitor scraping, never an invented persona) and produces the versioned voice record — a per-platform register map (incl. 小红书/微信公众号), banned phrases, per-context disclosure lines (the ECHO C2 upstream), a few-shot bank built exclusively from own posts, and 3-5 content pillars with Estimated %-allocations — submitted via memory/events/channels.ndjson for channel-registry to store as voice-dossier.md, the record every Craft-phase skill reads first. Not for the brand-level voice canon — use brand-language-codifier; not for audience/persona research — use audience-mapper. 声音档案/品牌语气/创始人语气/内容支柱/披露声明
Use when the user asks to "track our share of voice", "what share of the conversation do we own vs competitors", or "trend our SOV this quarter"; computes SOV% = brand mentions ÷ (brand + competitor panel mentions) per platform per period on a LOCKED competitor panel — a panel switch invalidates the trend (restart the series and log the break; the ECHO O3 rule) — plus a sentiment-weighted SOV variant (sentiment labeled Estimated unless human-coded) and a Wikipedia-pageviews attention-share alternative; built from keyless listening connectors, gdelt.py news echo, and user exports — public counts only, closed platforms are never scraped. Not for backlink or offsite SEO signals — use offsite-signal-analyzer. 声量份额/竞品声量对比/提及份额/注意力份额