The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted content from the web to detect changes based on natural language goals, creating a vulnerability surface for indirect prompt injection.
Ingestion points: External data enters the agent context via monitored targets specified by --page, --scrape-urls, --crawl-url, and --queries flags defined in SKILL.md.
Boundary markers: The skill does not provide instructions for using boundary markers or explicit delimiters to separate the monitored web content from the agent's instructions.
Capability inventory: The skill utilizes Bash(firecrawl *) and Bash(npx firecrawl-cli *) to manage monitors that can send data to external webhook-url targets or email addresses.
Sanitization: No sanitization, escaping, or validation of the monitored external content is described within the skill's instructions; processing logic is handled server-side by the vendor's AI judge.
[EXTERNAL_DOWNLOADS]: The skill instructions utilize npx firecrawl-cli in SKILL.md, which downloads the latest version of the vendor's utility from the NPM registry during execution.
[COMMAND_EXECUTION]: The skill executes shell commands using the firecrawl CLI to create, list, and manage web monitors. These commands involve network operations to communicate with the vendor's API and configuration of notification endpoints.

firecrawl-monitor