cogni-issues

Pass

Audited by Gen Agent Trust Hub on Jul 25, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill exhibits no malicious patterns, obfuscation, or attempts to bypass safety protocols.
  • [COMMAND_EXECUTION]: The skill executes local helper scripts and the gh CLI. Analysis of the scripts (gh-issues-helper.sh, issue-store.sh, resolve-plugin.sh) shows that shell commands are constructed safely using arguments rather than unsafe string interpolation or eval on user input.
  • [EXTERNAL_DOWNLOADS]: The skill requires the GitHub CLI as a prerequisite but does not download or execute unverified remote scripts. It provides installation hints for official sources.
  • [DATA_EXFILTRATION]: Communication is restricted to official GitHub API endpoints via the gh CLI. The skill does not handle user credentials directly, as authentication is managed by the CLI tool through a browser-based OAuth flow.
  • [PROMPT_INJECTION]: Instructions are instructional and scoped to the task. No patterns suggesting jailbreaks or system prompt extraction were found.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 25, 2026, 01:51 PM
Security Audit — agent-trust-hub — cogni-issues