bootstrap-issue-config

Pass

Audited by Gen Agent Trust Hub on Nov 19, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill makes extensive use of local CLI tools including the GitHub CLI (gh), git, and jq to perform repository analysis and metadata management. These are legitimate operations given the skill's purpose of automating repository setup.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests and processes data that could be controlled by an external actor.
  • Ingestion points: The skill reads existing labels, recent issue titles, issue templates, and contributor history from the repository metadata via gh and git commands (found in SKILL.md).
  • Boundary markers: There are no explicit instructions or delimiters used to ensure the agent ignores or sanitizes instructions that might be embedded within issue titles or label descriptions.
  • Capability inventory: The skill has the capability to write to the local file system (creating .github/issue-triage/config.json and .github/CODEOWNERS) and modify the remote repository state by creating new labels using gh label create.
  • Sanitization: While the skill uses jq to validate the syntax of the generated JSON configuration, it does not explicitly sanitize the individual strings harvested from the repository before they are interpolated into commands or written to configuration files.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 19, 2026, 01:12 AM