The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the shell to execute commands via npx @useclawlink/cli. This includes logging in, connecting to the Firecrawl service, and running various scraping/crawling actions. It notes that authentication credentials are stored locally at ~/.clawlink/credentials.json.
[EXTERNAL_DOWNLOADS]: The instructions direct the agent to use npx to download and run the @useclawlink/cli package from the official NPM registry at runtime.
[INDIRECT_PROMPT_INJECTION]: The skill's core functionality involves scraping, crawling, and searching external websites. This content is untrusted and is subsequently processed by the AI agent, creating a surface for indirect prompt injection attacks.
Ingestion points: Data returned from actions like firecrawl_scrape, firecrawl_crawl, and firecrawl_search (SKILL.md).
Boundary markers: None identified in the provided instructions to delimit scraped content.
Capability inventory: The agent has shell command execution capabilities through the CLI tool (SKILL.md).
Sanitization: No explicit sanitization or filtering of the scraped content is mentioned in the skill definition.

firecrawl