html-content-analysis
Pass
Audited by Gen Agent Trust Hub on Mar 9, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill documentation describes shell-based operations like
curl -s {url} | grep .... If the{url}input is not properly sanitized, it creates a risk of command injection where shell control characters could be used to execute arbitrary commands. - [EXTERNAL_DOWNLOADS]: The skill retrieves data from external URLs using
WebFetchandcurl. This functionality introduces a surface for Server-Side Request Forgery (SSRF) if the agent's network environment is not restricted. - [PROMPT_INJECTION]: The skill parses untrusted content (meta tags, comments, and structured data) from external websites, creating a vector for indirect prompt injection where a malicious site could influence the agent's behavior.
- [COMMAND_EXECUTION]: The skill configuration executes local hooks (
pre_network_skill_hook.shandpost_evidence_capture_hook.sh) via relative paths, which involves running scripts located outside the skill's own directory.
Audit Metadata