flow-next-opencode-prime

Warn

Audited by Gen Agent Trust Hub on Apr 7, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes arbitrary shell commands for project verification, including 'pytest --collect-only', 'npx jest --listTests', and 'pnpm build --help' (as defined in workflow.md). Executing these commands on an untrusted repository poses a risk, as project-specific configurations or malicious test suites could trigger unexpected code execution during the discovery or dry-run phase.- [EXTERNAL_DOWNLOADS]: The remediation templates (remediation.md) recommend installing pre-commit hooks from 'github.com/pre-commit/pre-commit-hooks' and 'github.com/astral-sh/ruff-pre-commit'. These are well-known and reputable sources for developer tooling and are used here to improve local feedback loops.- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted data from the repository being analyzed.
  • Ingestion points: Repository files and structure are scanned by multiple 'scout' tasks (e.g., 'tooling-scout', 'testing-scout', 'docs-gap-scout') as described in workflow.md.
  • Boundary markers: The skill lacks explicit delimiters or instructions to ignore embedded commands within the files it analyzes.
  • Capability inventory: The skill has the ability to execute shell commands, modify project configuration files (like 'package.json'), and create new documentation or environment files.
  • Sanitization: There is no evidence of sanitization or strict schema validation for the data ingested from the repository before it is used to generate scores and remediation recommendations.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 7, 2026, 09:07 AM