The Agent Skills Directory

[PROMPT_INJECTION]: Potential surface for indirect prompt injection. Ingestion points: The skill ingests backlink CSV files and domain URLs provided by users or external tools for analysis (SKILL.md). Boundary markers: No explicit delimiters or instructions are provided to the agent to treat external data as untrusted content. Capability inventory: The skill writes reports and summaries to the memory/monitoring/ directory and updates memory/open-loops.md (SKILL.md). Sanitization: The instructions do not define sanitization or validation protocols for the imported CSV data.
[EXTERNAL_DOWNLOADS]: The skill documentation includes installation commands using npx skills to fetch content from the vendor's remote repository. This aligns with the expected distribution model for the platform.
[NO_CODE]: The skill is implemented using markdown files and instructions only, with no executable code or scripts provided within the skill package.

backlink-analyzer