The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes content retrieved from the external VictoriaMetrics documentation site, creating an inherent surface for indirect prompt injection should the source documentation be modified with malicious instructions.
Ingestion points: The scripts/vmdocs.py script fetches documentation indices and page content from https://docs.victoriametrics.com/.
Boundary markers: While instructions guide the agent to quote and cite the link for every claim, there are no explicit delimiters or "ignore embedded instructions" warnings for the fetched text.
Capability inventory: The skill is restricted to Bash(python3:*) and Read tools. The script performs network fetches, local caching, and text extraction.
Sanitization: The script includes logic to filter out potentially hazardous HTML elements such as <script>, <style>, and <nav> tags, and unescapes HTML entities before returning the text.
[EXTERNAL_DOWNLOADS]: Fetches documentation indexes and content from the official VictoriaMetrics documentation domain (docs.victoriametrics.com). This network activity is documented neutrally as it aligns with the skill's primary purpose and targets the vendor's own infrastructure.

victoriametrics-docs