The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes untrusted data from the open web.
Ingestion points: SKILL.md (Step 3 and Step 4 use firecrawl_search, firecrawl_scrape, web_search_exa, and crawling_exa to fetch external content).
Boundary markers: Absent. The instructions do not include specific delimiters or directives to ignore instructions embedded within the retrieved web content.
Capability inventory: The skill can search the web, scrape content, and write findings to local files (Step 6). It does not exhibit higher-risk capabilities like arbitrary shell command execution or direct exfiltration of sensitive local files.
Sanitization: Absent. There is no mention of filtering or sanitizing the external data before it is synthesized into the final report.

deep-research