github-issues

Fail

Audited by Gen Agent Trust Hub on Apr 25, 2026

Risk Level: HIGHCREDENTIALS_UNSAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill attempts to harvest credentials by reading sensitive local files. Evidence: It specifically searches ~/.hermes/.env and ~/.git-credentials to extract GITHUB_TOKEN values using shell tools.\n- [COMMAND_EXECUTION]: Shell pipelines involving grep, sed, and cut are used to parse sensitive system files and extract secrets.\n- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection by ingesting untrusted data from GitHub and presenting it to the agent without sanitization.\n
  • Ingestion points: SKILL.md fetches issue titles and bodies from the GitHub API.\n
  • Boundary markers: No delimiters or warnings are used when displaying issue content to the agent.\n
  • Capability inventory: The skill has access to sensitive local files, shell command execution, and network operations.\n
  • Sanitization: No escaping or validation is performed on the data retrieved from the GitHub API.
Recommendations
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Apr 25, 2026, 12:08 PM
Security Audit — agent-trust-hub — github-issues