The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external, untrusted sources such as live search engine results pages (SERPs) and competitor websites during its research phase.
Ingestion points: Capability research workflows in references/capability-research.md (Google SERP Analysis, SimilarWeb Competitor Intelligence, Answer The Public keyword research).
Boundary markers: The instructions do not specify delimiters or safety warnings to ignore instructions embedded in the external content being analyzed.
Capability inventory: The skill utilizes agent-browser for automation and writes detailed reports and action plans to the local project directory (e.g., ./.pawbytes/marketing-suites/brands/).
Sanitization: There is no explicit requirement for the agent to sanitize or filter the retrieved web content before processing it.
[EXTERNAL_DOWNLOADS]: The skill references and provides installation instructions for an external browser automation tool from a well-known service.
Evidence: references/shared-patterns.md contains instructions to fetch the tool: npx skills add https://github.com/vercel-labs/agent-browser --skill agent-browser.
[COMMAND_EXECUTION]: The skill uses local shell commands to execute its research and technical audit functions.
Evidence: references/capability-research.md defines workflows using agent-browser CLI commands for SERP checking and schema validation. references/shared-patterns.md includes shell scripts (tool-discovery.sh, chrome-profiles.sh) for environmental setup.

paw-mkt-seo