The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to scrape content from external, untrusted URLs. An attacker who controls a target website could embed malicious instructions within the web content (e.g., hidden markdown or HTML comments) to manipulate the agent's subsequent actions.
Ingestion points: The firecrawl scrape command fetches data from arbitrary user-provided URLs.
Boundary markers: The skill does not explicitly define delimiters for the scraped content, though it recommends using bounded reads like head or grep for inspection.
Capability inventory: The skill allows the use of the Bash tool to execute CLI commands.
Sanitization: No specific sanitization or filtering of the scraped content is mentioned before the agent processes it.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute firecrawl and npx commands for scraping operations.
[EXTERNAL_DOWNLOADS]: The skill uses npx firecrawl-cli, which downloads the necessary command-line interface from the public NPM registry at runtime.

firecrawl-scrape