The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the firecrawl command-line tool and npx firecrawl-cli to execute monitoring tasks, including creating, listing, and running monitors. These are vendor-owned tools associated with the skill's primary function.
[INDIRECT_PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it ingests untrusted data from websites monitored by the user. This content is then processed by an AI judge which follows natural language instructions provided in the --goal and prompt parameters.
Ingestion points: Web content is fetched from URLs provided in --page, --crawl-url, or --scrape-urls flags in SKILL.md and json-tracking.md.
Boundary markers: No explicit delimiters or boundary markers are instructed for the content being monitored, increasing the risk that malicious instructions on a monitored page could influence the AI judge's behavior.
Capability inventory: The skill can trigger notifications to external webhooks (--webhook-url) and email addresses (--email).
Sanitization: There are no instructions for sanitizing or escaping the content retrieved from external URLs before it is evaluated by the AI judge.

firecrawl-monitor