code-explorer

Pass

Audited by Gen Agent Trust Hub on Jun 21, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill uses the WebFetch tool to retrieve code from external sources such as GitHub repositories based on user input for analysis.
  • [COMMAND_EXECUTION]: The skill requests access to the Bash tool in its configuration. While the current instructions focus on code analysis and explanation, the presence of this tool grants the agent the ability to execute shell commands on the local system.
  • [INDIRECT_PROMPT_INJECTION]: The skill represents an indirect prompt injection surface as it processes untrusted data from external codebases.
  • Ingestion points: External code fetched via WebFetch from user-provided URLs (SKILL.md).
  • Boundary markers: None explicitly defined in the instructions to separate fetched code from agent instructions.
  • Capability inventory: Bash, Read, Grep, WebFetch (SKILL.md frontmatter).
  • Sanitization: The skill relies on the agent's internal analysis and does not specify explicit validation or escaping of the fetched content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 21, 2026, 07:38 AM
Security Audit — agent-trust-hub — code-explorer