sync-github-epic

Pass

Audited by Gen Agent Trust Hub on Nov 10, 2026

Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill performs legitimate operations using local scripts and system utilities such as gh CLI, node, grep, and jq to interact with the GitHub API and local filesystem.
  • [DYNAMIC_EXECUTION]: Shell scripts utilize eval for variable expansion to ensure compatibility between different shell environments (Bash and Zsh); however, the variables are constructed from internal constants, mitigating the risk of arbitrary command injection.
  • [DATA_EXFILTRATION]: The skill accesses local project configuration files and Epic markdown documents to transmit their content to GitHub Issues as part of its primary synchronization function.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted markdown content from Epic files, which is a potential vector for indirect prompt injection; the risk is managed by structural formatting rules defined in tracker-card-summary.md.
  • [SAFE]: The skill demonstrates high security maturity by including defer-mutation.js, which implements comprehensive secret redaction and access control gates to prevent unauthorized or unrecorded mutations to external trackers.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 10, 2026, 03:29 PM