The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted data from arbitrary external websites, which serves as a surface for indirect prompt injection attacks. Malicious content on a scraped page could attempt to override agent instructions or exfiltrate context.
Ingestion points: Data enters the agent's context through the output of the firecrawl scrape command, which is saved to files or printed to stdout as markdown.
Boundary markers: The skill does not define specific boundary markers or instructions for the agent to ignore embedded commands within the scraped content.
Capability inventory: The skill environment permits shell command execution (Bash) and file system operations.
Sanitization: While the skill claims to return 'clean' markdown, there is no evidence of sanitization specifically targeting prompt injection attempts.
[EXTERNAL_DOWNLOADS]: The skill instructions include the use of npx firecrawl-cli, which involves downloading and executing the Firecrawl CLI package from the npm registry at runtime.
[COMMAND_EXECUTION]: The skill relies on shell commands that incorporate user-provided URLs. The skill includes a specific safety recommendation to 'Always quote URLs' to prevent the shell from interpreting special characters like ? and &, which mitigates potential command injection risks.

firecrawl-scrape