awesome-repositories.com
Blog
MCP
awesome-repositories.com

Discover the best open-source repositories with AI-powered search.

ExploreCurated searchesOpen-source alternativesSelf-hosted softwareBlogSitemap
ProjectMCP serverAboutHow we rankPress
LegalPrivacyTerms
© 2026 Bringes Technology SRL·VAT RO45896025·hello@awesome-repositories.com
microsoft avatar

microsoft/skill-recorder

0
View on GitHub↗

Skill Recorder

Skill Recorder is a desktop app that records an on-screen work session and turns it into a reusable skill or scheduled automation for AI agents. It captures clicks, app and window switches, visited pages, clipboard previews, terminal activity and spoken narration in the background, controlled by a global shortcut and a floating bar. It can also import an existing MP4 or WebM demonstration instead of recording live. Everything is recorded and stored locally until the user chooses to analyze it.

Analysis uses the GitHub Copilot SDK, with sign-in through the browser via enterprise single sign-on or a personal account. A multi-turn session reconstructs the recording into one overall intent and an ordered list of steps that can be reviewed, edited and reordered. Once a user approves the analysis, the app generates a skill file or a scheduled automation. The output favors native tools over replaying interface clicks and keeps calculation steps separate from steps that change things outside. A read-only preview of the generated skill text is available before installing or exporting.

Privacy handling runs on the device. Terminal commands and transcripts are scanned for secrets and personal data, and detected values are masked before any excerpt goes to a cloud service. Optional advanced protection uses text recognition to blur sensitive values in screenshots, including values split across several words. Narration is transcribed locally with a multilingual speech model, so audio is not sent over the network.

The repository also includes an offline evaluation harness. It scores inferred intents and steps against rubrics, checks generated skill plans and automation builders against frozen analyses, runs repeatable business-workflow scenarios, and measures redaction recall and precision without models or network access. Installation on Windows is scripted.

Features

  • Desktop Screen Recorders - Records an on-screen work session in a desktop app and converts it into a reusable skill or scheduled automation for AI agents.
  • Workflow Intent and Step Reconstruction - Analyzes a recorded session with an AI assistant to produce one overall intent and an ordered list of steps that can be reviewed, edited and reordered.
  • Sensitive Data Redaction - Masks detected secrets and personal values in text before anything reaches the cloud AI service, keeping prose, URLs, and hashes intact.
  • Copilot API Integrations - Uses the GitHub Copilot SDK in a multi-turn session to reconstruct a recording into one intent and editable ordered steps.

AI search

Explore more awesome repositories

Describe what you need in plain English — the AI ranks thousands of curated open-source projects by relevance.

Start searching with AI
5,710 stars·651 forks·TypeScript·MIT·1 view
  • Session Recorders - Records work sessions including clicks, app switches, clipboard previews, terminal activity and narration, or imports an existing video demonstration for analysis.
  • Screen Capture Utilities - Captures on-screen work including clicks, window switches, clipboard, terminal activity and narration in the background, or imports an existing demonstration video.
  • Skill Generation - Turns an approved analysis into a structured skill file or scheduled automation that prefers native tools and separates calculation from side-effecting steps.
  • Agent-Driven Task Automation - Turns a demonstrated knowledge-work task into a reusable agent skill or scheduled automation so repeated work needs no manual effort.
  • Secret and Personal Data Detection - Scans terminal commands and transcripts on the device for secrets and personal data, using validators to avoid flagging invalid numbers.
  • Sensitive Variable Redaction - Masks secrets and personal data in text and blurs sensitive values in screenshots on the device before anything is sent to the cloud.
  • Agent Skill Generators - Generates a skill file or scheduled automation from an approved analysis, favoring native tools over replaying interface clicks.
  • On-Device Speech Recognizers - Runs a multilingual speech recognition model locally on recorded narration, producing transcripts without sending audio over the network.
  • Speech-to-Text Transcribers - Converts recorded spoken commentary into text on the user's device with a multilingual speech model, without sending audio to the cloud.
  • On-Device Transcriptions - Transcribes recorded narration into text locally with a multilingual speech model, so audio is never sent over the network.
  • Digital Activity Recorders - Captures window switches, visited pages, clipboard previews, terminal activity, and narration to give each recorded step supporting context.
  • Recurring Task Automation - Packages a reconstructed procedure as a scheduled automation that runs again on a recurring timetable.
  • Automation Skills - Converts a demonstrated workflow into a reusable skill file that an agent can run on demand.
  • Session Analysis Integrations - Uses the GitHub Copilot SDK in a multi-turn session to reconstruct a recording into one intent and editable ordered steps.
  • Local-First Screen Captures - Captures screen activity, clipboard previews, terminal sessions and narration on the device, keeping all raw recording data local until the user chooses to analyze.
  • Automation Schedulers - Creates a scheduled automation for autonomous agents from an approved analysis so a recorded task can rerun without manual effort.
  • Privacy Blurring Tools - Uses text recognition to find sensitive values in recorded screenshots, including values split across words, and blurs exactly those regions.
  • Video Recording Imports - Imports an existing MP4 or WebM demonstration, validates its limits, and analyzes its frames and speech instead of recording live.
  • PII Recognition Pipelines - Runs an on-device pipeline that detects and masks secrets and personal data in text and blurs sensitive screen regions before any cloud analysis.
  • Privacy and Data Protection - Scans terminal commands and transcripts on the device for secrets and personal data before any excerpt is sent, keeping recording data local until analysis.
  • Prompt Redaction Gateways - Passes every outgoing text excerpt through on-device secret and personal-data detectors and masking so only sanitized content reaches a cloud service.
  • Screenshot Sensitive Value Blurring - Locates sensitive values in recorded frames with text recognition, including values split across words, and blurs those pixel regions before frames are used.
  • Agent Session Reconstruction - Runs a multi-turn model session that reconstructs recorded events and frames into one overall intent and an ordered list of steps.
  • Model Output Quality Evaluations - Scores inferred intents, steps, plans, and tool choices against rubrics and frozen scenarios offline, and measures redaction recall and precision, to expose regressions and variance.

    • Star history

    Star history chart for microsoft/skill-recorderStar history chart for microsoft/skill-recorder

    How this analysis was created: This summary and feature list are AI-generated from collected project material and can contain mistakes. Stars, license and language are imported from GitHub. Inclusion does not mean that we have tested or audited this project. Check the source documentation for any feature you depend on. Learn more on our About page.

    Projects sharing features with Skill Recorder

    These projects share indexed features with Skill Recorder. Shared tags can include platform or build tooling; verify the primary use case before treating a result as a replacement.
    • linuxhsj/openclaw-zero-tokenlinuxhsj avatar

      linuxhsj/openclaw-zero-token

      5,213View on GitHub↗

      OpenClaw Zero Token is a self-hosted gateway and assistant platform that lets people chat with browser-based AI services such as ChatGPT, Claude, Gemini, DeepSeek, Kimi and Grok without paying for API tokens. It captures a login from a debuggable Chrome browser through the Chrome DevTools Protocol, then reuses that session's cookies and headers for later requests. A single OpenAI-compatible chat completions endpoint, secured by a bearer token, routes each request to whichever web model is configured, so outside programs can call those models like a standard API. Because web chat models have n

      TypeScript
      View on GitHub↗5,213
    • maziyarpanahi/openmedmaziyarpanahi avatar

      maziyarpanahi/openmed

      5,785View on GitHub↗

      OpenMed is a local-first healthcare AI toolkit for clinical text. It finds diseases, drugs, anatomy, genes and personal identifiers in medical notes, then masks, replaces, hashes or date-shifts the identifiers under HIPAA, GDPR and research-style policy profiles. It also turns extracted entities into terminology-coded FHIR resources and OMOP records. A curated registry of specialized medical entity-recognition models, selectable by name, sits underneath all of this. Its de-identification work covers dozens of languages and scripts. It uses locale-aware fake replacements, checksum-valid nation

      Pythonandroidclinical-nlphealthcare
      View on GitHub↗5,785
    • microsoft/foundry-localmicrosoft avatar

      microsoft/Foundry-Local

      2,380View on GitHub↗

      Foundry-Local is a machine learning development tool designed to facilitate private, on-device inference and model management. It provides a local server environment that hosts machine learning models directly on the user's hardware, ensuring that all data processing, including prompt handling and audio transcription, remains within the local environment without requiring external cloud connectivity. The project distinguishes itself by automating the entire model lifecycle, including the discovery, downloading, and versioning of assets to maintain compatibility with host hardware. It features

      C++ai-sdkchat-completionsfoundry-local
      View on GitHub↗2,380
    • nickemanarin/screentogifNickeManarin avatar

      NickeManarin/ScreenToGif

      27,114View on GitHub↗

      ScreenToGif is a screen recording and GIF editor designed to capture screen activity and refine frames to produce GIF animations and video files. It functions as a screen-to-video capture tool for creating visual demonstrations, documentation, and tutorials. The application includes a non-linear frame editor that allows for the modification of recorded footage before export. It can capture a selected screen area, webcam feed, or sketchboard, and exports the resulting animations as images, videos, or GIFs. The tool supports user interface translation to provide a localized experience in multi

      C#editorgifrecorder
      View on GitHub↗27,114
    Compare all 30 related projects→

    Frequently asked questions

    What does microsoft/skill-recorder do?

    Skill Recorder is a desktop app that records an on-screen work session and turns it into a reusable skill or scheduled automation for AI agents. It captures clicks, app and window switches, visited pages, clipboard previews, terminal activity and spoken narration in the background, controlled by a global shortcut and a floating bar. It can also import an existing MP4 or WebM demonstration instead of recording live. Everything is recorded and stored locally until the user…

    What are the main features of microsoft/skill-recorder?

    The main features of microsoft/skill-recorder are: Desktop Screen Recorders, Workflow Intent and Step Reconstruction, Sensitive Data Redaction, Copilot API Integrations, Session Recorders, Screen Capture Utilities, Skill Generation, Agent-Driven Task Automation.

    Which projects share features with microsoft/skill-recorder?

    Projects with overlapping indexed features include: linuxhsj/openclaw-zero-token — OpenClaw Zero Token is a self-hosted gateway and assistant platform that lets people chat with browser-based AI… maziyarpanahi/openmed — OpenMed is a local-first healthcare AI toolkit for clinical text. It finds diseases, drugs, anatomy, genes and… microsoft/foundry-local — Foundry-Local is a machine learning development tool designed to facilitate private, on-device inference and model… nickemanarin/screentogif — ScreenToGif is a screen recording and GIF editor designed to capture screen activity and refine frames to produce GIF… mathewsachin/captura — Captura is a desktop screen recording and screenshot utility designed to capture video, webcam feeds, and system audio… yuka-friends/windrecorder — Windrecorder is a local activity logger and visual memory augmentation tool that captures screenshots and video…