github-profile

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the GitHub CLI (gh) and git to interact with the user's account. It fetches user details, repository lists, and contribution data. It also creates a new repository and pushes code to GitHub.
  • [EXTERNAL_DOWNLOADS]: The skill incorporates several well-known third-party services to generate visual elements for the profile, including img.shields.io for badges, skillicons.dev for tech stack icons, readme-typing-svg.demolab.com for animations, and komarev.com for view counters. These are standard industry tools for GitHub profile customization.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources (the user's own GitHub bio and repository descriptions) and processes it to generate the README content.
  • Ingestion points: Data is gathered from gh api user and gh repo list (specifically the bio, name, and description fields).
  • Boundary markers: The instructions do not explicitly require the agent to use delimiters or ignore potential instructions embedded within the user's GitHub metadata.
  • Capability inventory: The skill possesses the capability to create local files, create remote repositories, and execute git commands to push content to a public profile.
  • Sanitization: The instructions do not define a specific sanitization or filtering process for the text pulled from the user's profile before it is used to generate the 'About Me' section or project descriptions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 11:13 AM
Security Audit — agent-trust-hub — github-profile