gist-screenshot-evidence
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFE
Full Analysis
- [COMMAND_EXECUTION]: The skill interacts with the system using the gh and git command-line interfaces. Technical implementation via subprocess.run utilizes argument lists, ensuring commands are executed without shell interpretation, which effectively mitigates command injection risks.
- [DATA_EXFILTRATION]: To perform its primary function, the skill retrieves the user's GitHub authentication token. This token is used to authenticate Git operations and API calls to GitHub Gists. While this involves sensitive credential access, it is scoped to the skill's intended purpose of managing visual evidence on the user's GitHub account.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data from GitHub Gist metadata. To protect against potential injection via malicious gist descriptions or filenames, the skill implements regex-based filename sanitization and strict validation of paths, reducing the attack surface for data-driven exploits.
Audit Metadata