The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves fetching and processing untrusted content from external websites (search results and competitor pages), which presents a potential vulnerability for indirect prompt injection.
Ingestion points: The skill is instructed in SKILL.md (Step 1 and Step 5) to read the top 10 Google results and fetch/read the top 3 competitor pages.
Boundary markers: Absent; the instructions do not include any explicit delimiters or guidance to treat the fetched content as untrusted data separate from the agent's instructions.
Capability inventory: The agent uses the fetched data to generate strategic ranking plans, content gap analyses, and word count targets.
Sanitization: Absent; there are no instructions to sanitize or validate the content retrieved from external URLs before it is processed.
[EXTERNAL_DOWNLOADS]: The skill retrieves data from external search results and competitor URLs as part of its primary SEO research workflow.
Evidence: SKILL.md mandates fetching external content in Step 1 ("Google the target keyword. Read the top 10 results") and Step 5 ("Fetch and read the top 3 pages"). The reference files additionally point to various well-known SEO industry sources (Ahrefs, Semrush, Backlinko) for benchmarking data.

keyword-deep-dive