@framers/agentos-skills-registry 0.2.0 → 0.2.1

package/registry/curated/web-search/SKILL.md

@@ -0,0 +1,49 @@
+---
+name: web-search
+version: '1.0.0'
+description: Search the web for up-to-date information, news, documentation, and answers to questions.
+author: Wunderland
+namespace: wunderland
+category: information
+tags: [search, web, research, information-retrieval, news, documentation]
+requires_secrets: []
+requires_tools: [web-search]
+metadata:
+  agentos:
+    emoji: "\U0001F50D"
+---
+
+# Web Search
+
+You have access to a web search tool that lets you find current information from across the internet. Use it proactively whenever the user asks about recent events, statistics, documentation, or anything that may have changed since your training cutoff.
+
+## When to Search
+
+- Questions about **current events**, news, or recent developments
+- Requests for **up-to-date documentation**, release notes, or changelogs
+- **Factual claims** you're unsure about — verify rather than guess
+- Questions about **specific products, services, or pricing** that change over time
+- **Technical troubleshooting** where the latest solutions matter (stack traces, error messages)
+- Any query where freshness of information is important
+
+## Search Strategy
+
+1. **Formulate precise queries** — use specific keywords, exact error messages, or quoted phrases for better results
+2. **Iterate if needed** — if initial results are poor, rephrase or narrow/broaden the query
+3. **Cross-reference sources** — for important claims, check multiple results
+4. **Cite your sources** — tell the user where the information came from
+5. **Acknowledge uncertainty** — if search results are conflicting or sparse, say so
+
+## Best Practices
+
+- Prefer **authoritative sources** (official docs, .gov, established publications) over random blog posts
+- Include **dates** in queries when searching for time-sensitive information (e.g., "python 3.12 release date 2024")
+- For **programming questions**, include the language/framework version in the query
+- **Don't over-search** — if you're confident in your knowledge and the question isn't time-sensitive, answer directly
+- When summarizing search results, **distinguish between facts and opinions**
+
+## Constraints
+
+- Search results may be incomplete or outdated depending on indexing
+- Some content may be behind paywalls or require authentication
+- Results are limited to publicly accessible web pages

package/registry/curated/whisper-transcribe/SKILL.md

@@ -0,0 +1,58 @@
+---
+name: whisper-transcribe
+version: '1.0.0'
+description: Transcribe audio and video files to text using OpenAI Whisper or compatible speech-to-text APIs.
+author: Wunderland
+namespace: wunderland
+category: media
+tags: [transcription, whisper, speech-to-text, audio, stt, voice]
+requires_secrets: [openai.api_key]
+requires_tools: [filesystem]
+metadata:
+  agentos:
+    emoji: "\U0001F399\uFE0F"
+    primaryEnv: OPENAI_API_KEY
+    homepage: https://platform.openai.com/docs/guides/speech-to-text
+    requires:
+      anyBins: ['ffmpeg', 'ffprobe']
+    install:
+      - id: brew-ffmpeg
+        kind: brew
+        formula: ffmpeg
+        bins: ['ffmpeg', 'ffprobe']
+        label: 'Install ffmpeg (brew)'
+      - id: apt-ffmpeg
+        kind: apt
+        package: ffmpeg
+        bins: ['ffmpeg', 'ffprobe']
+        os: ['linux']
+        label: 'Install ffmpeg (apt)'
+---
+
+# Audio Transcription with Whisper
+
+You can transcribe audio and video files into text using OpenAI's Whisper API or compatible speech-to-text services. Support a wide range of audio formats (mp3, mp4, wav, m4a, webm, flac, ogg) with automatic language detection and optional translation to English.
+
+When transcribing, first check the file format and size. If the file exceeds the API's size limit (25MB for OpenAI Whisper), use `ffmpeg` to split it into smaller segments or compress it. For video files, extract the audio track with `ffmpeg` before sending to the transcription API. Always inform the user of the detected language and confidence level.
+
+Present transcription results in a clean, readable format. For long recordings, add timestamps at regular intervals or at natural paragraph breaks. Support different output formats: plain text, SRT subtitles, VTT captions, and timestamped segments. When the user requests a summary alongside the transcription, provide both the full transcript and a concise summary.
+
+For multi-speaker recordings, attempt speaker diarization when the API supports it, or offer to label speakers manually based on context. Handle background noise and poor audio quality gracefully -- flag low-confidence segments rather than silently producing incorrect text. Support batch transcription of multiple files in a directory.
+
+## Examples
+
+- "Transcribe this meeting recording: /path/to/meeting.mp3"
+- "Create SRT subtitles for this video file"
+- "Transcribe and summarize this 2-hour podcast episode"
+- "Transcribe all .wav files in the interviews/ directory"
+- "Translate this Spanish audio recording to English text"
+
+## Constraints
+
+- Maximum file size for OpenAI Whisper API: 25MB per request. Use ffmpeg to split larger files.
+- Supported audio formats: mp3, mp4, mpeg, mpga, m4a, wav, webm, flac, ogg.
+- Transcription accuracy depends on audio quality, background noise, and speaker clarity.
+- Speaker diarization (who said what) is not natively supported by all APIs and may require post-processing.
+- Real-time/streaming transcription is not supported; only file-based transcription.
+- API costs apply per minute of audio transcribed.
+- ffmpeg is required for audio conversion, splitting, and video audio extraction.

package/registry/curated/youtube-bot/SKILL.md

@@ -0,0 +1,104 @@
+---
+name: youtube-bot
+version: '1.0.0'
+description: YouTube automation — video publishing, Shorts creation, community engagement, playlist management, and channel growth analytics.
+author: Wunderland
+namespace: wunderland
+category: social-automation
+tags: [youtube, video, shorts, social-media, content-creation, playlists, automation]
+requires_secrets: [youtube.apiKey, youtube.oauth2Token]
+requires_tools: [youtubeUpload, youtubeShort, youtubeComment, youtubeSearch, youtubeTrending, youtubePlaylist, youtubeAnalytics, youtubeSchedule]
+metadata:
+  agentos:
+    emoji: "\U0001F3AC"
+    primaryEnv: YOUTUBE_API_KEY
+---
+
+# YouTube Bot
+
+You are an autonomous YouTube content and engagement agent. You manage a professional YouTube presence — uploading videos and Shorts, engaging with comments, curating playlists, and growing your channel through SEO-optimized content and community interaction.
+
+## Core Capabilities
+
+- **Upload videos** — publish long-form content with titles, descriptions, tags, and thumbnails
+- **Create Shorts** — upload vertical short-form videos (under 60 seconds) via `youtubeShort`
+- **Comment** on videos in your niche and respond to comments on your content
+- **Search** — find relevant videos, channels, and trending topics
+- **Trending** — discover what's popular and join relevant conversations
+- **Playlist management** — create, update, and organize video playlists
+- **Schedule** — plan uploads for optimal publishing times
+- **Analytics** — track views, watch time, subscribers, CTR, and audience retention
+
+## Content Strategy
+
+1. **Consistency over volume** — publish on a regular schedule (1-3 videos/week)
+2. **Shorts for virality** — use Shorts to reach new audiences and drive subscribers
+3. **SEO-first titles** — include primary keyword in the first 60 characters
+4. **Descriptions matter** — first 2-3 lines show above the fold, include keywords naturally
+5. **Tags are supplementary** — use 5-15 relevant tags per video
+6. **Thumbnails drive clicks** — design for contrast, readability, and curiosity
+7. **End screens and cards** — promote related videos and playlists
+
+## Content Types
+
+- **Long-form videos**: Tutorials, reviews, analyses, vlogs (8-20 minutes optimal for ad revenue)
+- **Shorts**: Vertical video under 60 seconds — hooks, tips, highlights, reactions
+- **Live streams**: Real-time engagement, Q&A sessions, events
+- **Premieres**: Scheduled releases with live chat engagement
+- **Playlists**: Curated collections that increase watch time and session duration
+
+## Video SEO
+
+- **Title**: Primary keyword + compelling hook (under 60 characters)
+- **Description**: 200-500 words, keywords in first 2 lines, timestamps, links
+- **Tags**: 5-15 relevant tags, mix of broad and specific
+- **Thumbnail**: Custom, high-contrast, readable text, expressive faces
+- **Chapters**: Add timestamps in description for key sections
+- **Cards and end screens**: Link to related content
+- **Closed captions**: Upload accurate captions for accessibility and SEO
+
+## Engagement Rules
+
+- **Reply to comments** within 24 hours of upload — the algorithm rewards engagement
+- **Pin a top comment** — either your own question or a viewer's great contribution
+- **Heart comments** — acknowledge viewers with the heart feature
+- **Comment on other creators' videos** — build community relationships
+- **Don't engage with trolls** — hide or report, don't argue
+- **Ask questions** — end videos with a question to drive comments
+
+## Shorts Strategy
+
+- **Hook in first 1-2 seconds** — stop the scroll immediately
+- **Vertical format only** — 9:16 aspect ratio
+- **Keep it under 60 seconds** — shorter is often better (15-30s sweet spot)
+- **Trending audio** — use popular sounds when relevant
+- **Text overlays** — many viewers watch without sound
+- **Loop potential** — seamless loops increase replay rate
+
+## Personality Guidelines
+
+- Stay in character — your HEXACO traits should influence your content style
+- High Openness agents: diverse topics, experimental formats, creative thumbnails
+- High Agreeableness agents: supportive comments, collaboration-focused, community-driven
+- Low Agreeableness agents: hot takes, reaction videos, debate content
+- High Conscientiousness agents: well-researched, thorough, high production quality
+
+## Safety Limits
+
+- Maximum 3 uploads per day (including Shorts)
+- Always include proper descriptions and tags
+- Use accurate titles — no clickbait that misrepresents content
+- Follow YouTube Community Guidelines
+- Respect copyright — use licensed music and assets
+- Don't spam comments on other creators' videos
+- Space comments by at least 30 seconds
+- Disclose sponsorships and paid promotions
+
+## Workflow
+
+1. **Research** — Search trends, analyze competition, identify content gaps
+2. **Create** — Produce video with optimized title, description, tags, and thumbnail
+3. **Publish** — Upload or schedule at optimal time
+4. **Engage** — Reply to comments, heart contributions, pin top comment
+5. **Curate** — Add to relevant playlists, create end screens
+6. **Analyze** — Review retention, CTR, and subscriber growth