aiden-runtime 3.19.5 → 3.19.7

package/workspace-templates/skills/web_research/SKILL.md

@@ -0,0 +1,18 @@
+---
+name: web_research
+description: Deep web research with multi-source verification
+version: 1.0.0
+tags: web, research, search, analyze
+license: Apache-2.0
+---
+
+# Web Research
+
+When performing web research:
+1. Search broad first, then narrow by entity
+2. Fetch actual page content, not just snippets
+3. Cross-reference at least 2 sources before stating facts
+4. Extract specific data points: names, numbers, dates, comparisons
+5. Never state information you cannot verify from the fetched content
+6. For comparisons: create structured tables with consistent attributes
+7. Always note the source URL for key facts

package/workspace-templates/skills/web_research/skill.json

@@ -0,0 +1,20 @@
+{
+  "name": "web_research",
+  "version": "1.0.0",
+  "description": "Deep web research with multi-source verification",
+  "author": "local",
+  "license": "MIT",
+  "tools": [],
+  "trigger_phrases": [],
+  "compatible_agents": [
+    "aiden"
+  ],
+  "min_agent_version": "3.0.0",
+  "tags": [
+    "web",
+    "research",
+    "search",
+    "analyze"
+  ],
+  "created": "2026-04-27T17:11:41.095Z"
+}

package/workspace-templates/skills/xitter/SKILL.md

@@ -0,0 +1,148 @@
+---
+name: xitter
+description: Post, read, and search X (Twitter) content using x-cli or the X API v2
+category: social
+version: 1.0.0
+origin: aiden
+license: Apache-2.0
+tags: twitter, x, social-media, tweet, post, search, api, x-cli, timeline, mentions
+---
+
+# X (Twitter) Automation
+
+Read, post, and search X content using the `x-cli` command-line tool or the X API v2 directly. Requires X API credentials.
+
+## When to Use
+
+- User wants to post a tweet or thread on X
+- User wants to search for recent tweets on a topic
+- User wants to read their home timeline or mentions
+- User wants to fetch tweets from a specific account
+- User wants to schedule or automate social media posts
+
+## How to Use
+
+### 1. Set up X API credentials
+
+Register at https://developer.twitter.com → Create Project → Create App. Get:
+- `API_KEY`, `API_SECRET`
+- `ACCESS_TOKEN`, `ACCESS_TOKEN_SECRET`
+- `BEARER_TOKEN` (for read-only app-only auth)
+
+```powershell
+$env:X_BEARER_TOKEN      = "your_bearer_token"
+$env:X_API_KEY           = "your_api_key"
+$env:X_API_SECRET        = "your_api_secret"
+$env:X_ACCESS_TOKEN      = "your_access_token"
+$env:X_ACCESS_SECRET     = "your_access_secret"
+```
+
+### 2. Search recent tweets (Bearer Token — no user auth needed)
+
+```python
+import requests, os
+
+def search_tweets(query, max_results=10):
+  headers = {"Authorization": f"Bearer {os.environ['X_BEARER_TOKEN']}"}
+  params  = {
+    "query":        f"{query} -is:retweet lang:en",
+    "max_results":  max_results,
+    "tweet.fields": "created_at,author_id,public_metrics"
+  }
+  resp = requests.get("https://api.twitter.com/2/tweets/search/recent", headers=headers, params=params)
+  resp.raise_for_status()
+  for tweet in resp.json().get("data", []):
+    print(f"• {tweet['text'][:120]}")
+    print(f"  {tweet['created_at']}  likes={tweet['public_metrics']['like_count']}\n")
+
+search_tweets("Aiden AI agent 2026", max_results=5)
+```
+
+### 3. Post a tweet (OAuth 1.0a)
+
+```python
+import requests, os, time, hashlib, hmac, base64, urllib.parse, uuid
+
+def post_tweet(text):
+  url    = "https://api.twitter.com/2/tweets"
+  params = {"text": text}
+
+  # OAuth 1.0a signature
+  oauth_params = {
+    "oauth_consumer_key":     os.environ["X_API_KEY"],
+    "oauth_nonce":            uuid.uuid4().hex,
+    "oauth_signature_method": "HMAC-SHA1",
+    "oauth_timestamp":        str(int(time.time())),
+    "oauth_token":            os.environ["X_ACCESS_TOKEN"],
+    "oauth_version":          "1.0",
+  }
+  signing_key  = "&".join([urllib.parse.quote(os.environ["X_API_SECRET"], safe=""),
+                            urllib.parse.quote(os.environ["X_ACCESS_SECRET"], safe="")])
+  base_string  = "&".join(["POST", urllib.parse.quote(url, safe=""),
+                            urllib.parse.quote("&".join(f"{k}={v}" for k,v in sorted(oauth_params.items())), safe="")])
+  signature    = base64.b64encode(hmac.new(signing_key.encode(), base_string.encode(), hashlib.sha1).digest()).decode()
+  oauth_params["oauth_signature"] = signature
+
+  auth_header  = "OAuth " + ", ".join(f'{k}="{urllib.parse.quote(v, safe="")}"' for k,v in oauth_params.items())
+  resp = requests.post(url, json=params, headers={"Authorization": auth_header, "Content-Type": "application/json"})
+  resp.raise_for_status()
+  print(f"Posted tweet: {resp.json()['data']['id']}")
+
+post_tweet("Hello from Aiden! 🤖 #AI #automation")
+```
+
+### 4. Fetch user timeline
+
+```python
+import requests, os
+
+def get_user_timeline(username, max_results=10):
+  headers = {"Authorization": f"Bearer {os.environ['X_BEARER_TOKEN']}"}
+  user    = requests.get(f"https://api.twitter.com/2/users/by/username/{username}", headers=headers).json()
+  user_id = user["data"]["id"]
+
+  resp    = requests.get(
+    f"https://api.twitter.com/2/users/{user_id}/tweets",
+    headers=headers,
+    params={"max_results": max_results, "tweet.fields": "created_at,public_metrics", "exclude": "retweets"}
+  )
+  for tweet in resp.json().get("data", []):
+    print(f"• {tweet['text'][:120]}\n  {tweet['created_at']}\n")
+
+get_user_timeline("elonmusk", max_results=5)
+```
+
+### 5. Read mentions
+
+```python
+import requests, os
+
+def get_mentions(user_id, max_results=10):
+  headers = {"Authorization": f"Bearer {os.environ['X_BEARER_TOKEN']}"}
+  resp    = requests.get(
+    f"https://api.twitter.com/2/users/{user_id}/mentions",
+    headers=headers,
+    params={"max_results": max_results, "tweet.fields": "created_at,author_id"}
+  )
+  for tweet in resp.json().get("data", []):
+    print(f"• {tweet['text'][:120]}\n  {tweet['created_at']}\n")
+```
+
+## Examples
+
+**"Search for recent tweets about LLM agents"**
+→ Use step 2 with query `"LLM agents"`.
+
+**"Post a tweet saying 'Aiden just automated my morning report'"**
+→ Ask user to confirm before posting, then use step 3.
+
+**"Show me the last 10 tweets from a specific account"**
+→ Use step 4 with the target username.
+
+## Cautions
+
+- Posting tweets requires user-context OAuth 1.0a — Bearer Token alone cannot post
+- X API free tier limits search results to the last 7 days and 1 request per 15 minutes
+- Always ask user for explicit confirmation before posting any tweet on their behalf
+- X API rate limits are strict — cache results locally rather than re-fetching on every call
+- Never hardcode API credentials — always use environment variables

package/workspace-templates/skills/xitter/skill.json

@@ -0,0 +1,26 @@
+{
+  "name": "xitter",
+  "version": "1.0.0",
+  "description": "Post, read, and search X (Twitter) content using x-cli or the X API v2",
+  "author": "aiden",
+  "license": "MIT",
+  "tools": [],
+  "trigger_phrases": [],
+  "compatible_agents": [
+    "aiden"
+  ],
+  "min_agent_version": "3.0.0",
+  "tags": [
+    "twitter",
+    "x",
+    "social-media",
+    "tweet",
+    "post",
+    "search",
+    "api",
+    "x-cli",
+    "timeline",
+    "mentions"
+  ],
+  "created": "2026-04-27T17:11:41.209Z"
+}

package/workspace-templates/skills/youtube-content/SKILL.md

@@ -0,0 +1,121 @@
+---
+name: youtube-content
+description: Extract transcripts, download audio/video, and analyze YouTube content using youtube-transcript-api and yt-dlp
+category: research
+version: 1.0.0
+origin: aiden
+license: Apache-2.0
+tags: youtube, transcript, video, audio, download, subtitles, yt-dlp, content, summary
+---
+
+# YouTube Content Extraction
+
+Extract transcripts, download audio or video, and retrieve metadata from YouTube using `youtube-transcript-api` (Python) and `yt-dlp`.
+
+## When to Use
+
+- User wants to get the transcript of a YouTube video
+- User wants to summarize a YouTube video without watching it
+- User wants to download audio from a YouTube video
+- User wants to download a video for offline viewing
+- User wants to search for metadata (title, duration, views) of a video
+
+## How to Use
+
+### 1. Install required tools
+
+```powershell
+pip install youtube-transcript-api
+pip install yt-dlp
+# Verify
+yt-dlp --version
+```
+
+### 2. Get a video transcript (Python)
+
+```python
+from youtube_transcript_api import YouTubeTranscriptApi
+
+video_id = "dQw4w9WgXcQ"   # from youtube.com/watch?v=<id>
+transcript = YouTubeTranscriptApi.get_transcript(video_id)
+text = " ".join(entry["text"] for entry in transcript)
+print(text[:2000])
+```
+
+### 3. Get transcript in a specific language
+
+```python
+from youtube_transcript_api import YouTubeTranscriptApi
+
+video_id = "dQw4w9WgXcQ"
+# Try English first, then auto-generated
+transcript = YouTubeTranscriptApi.get_transcript(video_id, languages=["en", "en-US"])
+text = " ".join(entry["text"] for entry in transcript)
+```
+
+### 4. List available transcript languages for a video
+
+```python
+from youtube_transcript_api import YouTubeTranscriptApi
+
+video_id = "dQw4w9WgXcQ"
+transcript_list = YouTubeTranscriptApi.list_transcripts(video_id)
+for t in transcript_list:
+  print(t.language, t.language_code, "auto-generated:", t.is_generated)
+```
+
+### 5. Download audio (MP3) with yt-dlp
+
+```powershell
+$videoUrl = "https://www.youtube.com/watch?v=dQw4w9WgXcQ"
+yt-dlp -x --audio-format mp3 -o "%(title)s.%(ext)s" $videoUrl
+```
+
+### 6. Download video (best quality)
+
+```powershell
+$videoUrl = "https://www.youtube.com/watch?v=dQw4w9WgXcQ"
+yt-dlp -f "bestvideo+bestaudio" --merge-output-format mp4 -o "%(title)s.%(ext)s" $videoUrl
+```
+
+### 7. Download video at specific quality
+
+```powershell
+# List available formats first
+yt-dlp -F $videoUrl
+
+# Download 720p
+yt-dlp -f "bestvideo[height<=720]+bestaudio" --merge-output-format mp4 -o "%(title)s.%(ext)s" $videoUrl
+```
+
+### 8. Get video metadata (no download)
+
+```powershell
+yt-dlp --dump-json --no-download $videoUrl | python -m json.tool | Select-String -Pattern '"title"|"duration"|"view_count"|"upload_date"'
+```
+
+### 9. Download an entire playlist
+
+```powershell
+$playlistUrl = "https://www.youtube.com/playlist?list=PLxxxxxx"
+yt-dlp -x --audio-format mp3 -o "%(playlist_index)s-%(title)s.%(ext)s" $playlistUrl
+```
+
+## Examples
+
+**"Get me the transcript of this YouTube tutorial so I can read it"**
+→ Use step 2 — extract video ID from URL and call `get_transcript`.
+
+**"Download the audio from this podcast episode on YouTube"**
+→ Use step 5 with the video URL to download as MP3.
+
+**"What languages are available for transcripts on this video?"**
+→ Use step 4 to list available transcript languages.
+
+## Cautions
+
+- `youtube-transcript-api` only works for videos that have transcripts (manual or auto-generated)
+- Auto-generated transcripts may have errors, especially for technical content or non-English speech
+- Downloading videos may be subject to YouTube's Terms of Service — use for personal research and fair use only
+- yt-dlp may need updates when YouTube changes its API: run `yt-dlp -U` to update
+- Rate limiting may occur on rapid consecutive downloads — add delays between requests for playlists

package/workspace-templates/skills/youtube-content/skill.json

@@ -0,0 +1,25 @@
+{
+  "name": "youtube-content",
+  "version": "1.0.0",
+  "description": "Extract transcripts, download audio/video, and analyze YouTube content using youtube-transcript-api and yt-dlp",
+  "author": "aiden",
+  "license": "MIT",
+  "tools": [],
+  "trigger_phrases": [],
+  "compatible_agents": [
+    "aiden"
+  ],
+  "min_agent_version": "3.0.0",
+  "tags": [
+    "youtube",
+    "transcript",
+    "video",
+    "audio",
+    "download",
+    "subtitles",
+    "yt-dlp",
+    "content",
+    "summary"
+  ],
+  "created": "2026-04-27T17:11:41.233Z"
+}