The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from external websites, which could be weaponized with malicious instructions to manipulate the agent's behavior during the audit.
Ingestion points: The agent fetches data from [domain]/robots.txt, page HTML (meta tags), and HTTP response headers (SKILL.md, Steps 1-3).
Boundary markers: No specific delimiters or instructions are provided to the agent to treat external content as untrusted data or to ignore embedded commands.
Capability inventory: The skill has access to Bash, WebFetch, and Write tools, allowing it to execute network commands and write files based on processed input.
Sanitization: There is no mention of sanitizing or escaping the fetched content before it is parsed or included in the final report.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute curl commands for retrieving remote files and site metadata (SKILL.md, Example section).
[DATA_EXFILTRATION]: The skill performs network operations using WebFetch and Bash to interact with external, non-whitelisted domains provided by the user (SKILL.md, Step 1).

geo-crawlers