The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill performs network requests to retrieve content from well-known services, including YouTube, TikTok, and Twitter/X (via the community-standard fxtwitter.com API). These operations are essential to the tool's primary purpose.
[DATA_EXFILTRATION]: The skill includes functionality to read local files via the fetch_local function. While this allows the agent to access local data such as PDFs and markdown files, it is the intended functionality of a file-fetching tool and does not involve exfiltration to unauthorized third-party servers.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface because it processes untrusted content from the web (transcripts, captions, and articles). Maliciously crafted external content could attempt to influence the agent's behavior.
Ingestion points: scripts/fetch.py (via trafilatura, yt-dlp, and youtube_transcript_api)
Boundary markers: The output uses YAML front-matter delimiters (---) to separate metadata from content, though it lacks a specific warning for the agent to ignore instructions within the fetched text.
Capability inventory: The skill can read local files and perform network GET requests; it does not have write access or the ability to execute arbitrary commands.
Sanitization: The script extracts plain text and normalizes whitespace but does not perform content-based filtering for injection patterns.

fetch-content