The Agent Skills Directory

[PROMPT_INJECTION]: The skill ingests and processes content from external URLs provided by the user, which creates a surface for indirect prompt injection attacks.
Ingestion points: Target URLs are processed using the WebFetch tool to conduct SEO audits as described in SKILL.md and defined in schemas/input.schema.json.
Boundary markers: The instructions lack specific guidance for the agent to use delimiters or ignore instructions embedded within the fetched web content.
Capability inventory: The skill possesses capabilities to write to the file system, execute shell commands, and perform further network operations (Write, Edit, Bash, WebFetch, WebSearch).
Sanitization: No sanitization or validation protocols are defined for handling external data before it is integrated into the agent's context.
[EXTERNAL_DOWNLOADS]: The skill utilizes npx to execute the Google-developed Lighthouse auditing tool, which may trigger a package download during execution.
Evidence: SKILL.md provides example commands using npx lighthouse to analyze website performance and SEO metrics.
[COMMAND_EXECUTION]: The skill uses the Bash tool to perform technical diagnostics on target domains.
Evidence: SKILL.md includes instructions for executing curl to retrieve robots.txt and sitemap.xml files, as well as running the Lighthouse CLI tool for site audits.

seo-optimization