The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external search engine sources, which presents a surface for indirect prompt injection attacks.
Ingestion points: Data is ingested from tools like get_ranked_keywords, get_serp_results, get_backlinks_overview, and search_local_businesses as described in SKILL.md.
Boundary markers: There are no explicit instructions for using delimiters or warnings to separate external data from system instructions.
Capability inventory: The skill can write results back to the application environment using update_project_context and generate reports via the seo-report skill.
Sanitization: No sanitization or validation logic is defined for the content fetched from the web or search result metadata.
[EXTERNAL_DOWNLOADS]: The skill references documentation and attribution links from its official service domain.
Evidence: References https://openseo.so/docs/skills/competitor-analysis for skill documentation and reporting attribution in the output format section of SKILL.md.

competitor-analysis