reference-seek

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches raw source code from GitHub via raw.githubusercontent.com to provide implementation examples to the user. This domain is a well-known service and the operation is central to the skill's purpose.
  • [COMMAND_EXECUTION]: The skill executes shell commands using the GitHub CLI (gh api) to search for repositories and retrieve file trees. It also uses curl to download file contents. These commands are used to implement the core search and retrieval functionality.
  • [PROMPT_INJECTION]: The skill has an indirect prompt injection surface because it ingests and displays code from arbitrary GitHub repositories based on user-provided topics.
  • Ingestion points: Raw code is fetched from GitHub repositories using curl in Step 3.
  • Boundary markers: External content is displayed within markdown code blocks in the synthesis step.
  • Capability inventory: The skill can execute shell commands (gh, curl) and use an internal Explore subagent to search the local codebase.
  • Sanitization: There is no explicit sanitization or filtering of the external code content before it is processed or presented to the user.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 01:29 PM
Security Audit — agent-trust-hub — reference-seek