file-ticket

Pass

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests issue templates from external repositories to structure its output, creating a surface for indirect prompt injection.
  • Ingestion points: The skill fetches raw content from .github/ISSUE_TEMPLATE/ in target repositories via the GitHub API in operations/github.md.
  • Boundary markers: Instructions in SKILL.md do not include explicit boundary markers or warnings to disregard instructions embedded within the fetched templates.
  • Capability inventory: The skill possesses write capabilities through the gh issue create command and read capabilities via gh api.
  • Sanitization: No sanitization or validation of the remote template content is specified before it is processed by the AI agent.
  • [DYNAMIC_EXECUTION]: The skill assembles shell commands by interpolating variables derived from the conversation context and user input.
  • Evidence: In operations/github.md, the create-ticket and search-duplicates operations build commands like gh issue create --title "" and gh issue list --search "".
  • Risk: If the values for , , or are not properly escaped by the execution environment, they could potentially be used to inject additional shell flags or parameters.
  • [COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (gh) to perform repository operations, including viewing repository details, listing labels, searching issues, and creating tickets.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 30, 2026, 01:35 PM
Security Audit — agent-trust-hub — file-ticket