Skip to content

Security: tjsasakifln/devflow

Security

SECURITY.md

Security Policy

Devflow is a local-first AI coding governance CLI. It scans code locally, generates reports, and never sends your code to any external service. This policy covers security vulnerabilities in the Devflow project itself.


Reporting a Vulnerability

If you discover a security vulnerability in Devflow, please report it through one of the following channels:

  1. GitHub Security Advisory: Create a private advisory
  2. Email: Reach out to the repository maintainers via the contact information in the GitHub profile

We ask that you do not report security vulnerabilities through public GitHub issues, discussions, or pull requests.

What to Include

  • A clear description of the vulnerability
  • Steps to reproduce (PoC or minimal example preferred)
  • Affected versions (check with devflow --version)
  • Potential impact
  • Any suggested remediation (optional)

Scope

In Scope

  • src/ -- all source code, including core logic, CLI, adapters, renderers, and kernel
  • templates/ -- project templates shipped with the CLI
  • action.yml -- the GitHub Action composite entry point
  • Dependency vulnerabilities in production dependencies (commander, @clack/prompts, picocolors)

Out of Scope

  • docs/ and examples/ -- documentation and example files
  • Third-party AI tools (Claude Code, Cursor, Copilot) -- vulnerabilities in those tools should be reported to their respective maintainers
  • Code generated or audited by Devflow -- Devflow is a governance tool, not a security guarantee
  • CI/CD pipeline configurations in consuming projects
  • Theoretical attacks requiring physical access, social engineering, or modified dependencies

Safe Harbor

We consider security research conducted in accordance with this policy as authorized conduct. We will not pursue legal action against researchers who:

  • Report vulnerabilities in good faith
  • Follow this disclosure policy
  • Do not access or modify data beyond what is necessary to demonstrate the vulnerability
  • Do not exploit a vulnerability beyond minimal proof of concept

Known Limitations

Devflow is a CLI tool, not a sandbox. The following limitations are acknowledged:

  • No runtime isolation: Devflow runs in the user's Node.js process. It cannot prevent deliberate bypass of its checks by a user who runs --force or modifies .devflow/ state files directly.
  • Heuristic checks, not formal verification: Pattern-based dangerous code detection has inherent false positive and false negative rates. It is a heuristic, not a formal proof of correctness or security.
  • No SAST/DAST integration yet: Static and dynamic application security testing are future work. Devflow does not currently replace dedicated security scanning tools.
  • Dependency scanning: Devflow does not scan for vulnerable dependencies. Use npm audit, snyk, or Dependabot for that purpose.
  • Token and secret detection: The built-in pattern check for hardcoded secrets is basic. For production secret scanning, use dedicated tools like truffleHog or git-secrets.

Dependencies

Devflow has minimal production dependencies:

Package Version Purpose
commander ^12.1 CLI argument parsing
@clack/prompts ^0.7 Interactive prompts
picocolors ^1.1 Terminal color output

Supply chain risk is low -- all packages are well-maintained, widely used, and have small surface areas. DevDependencies (TypeScript, vitest, tsx, rimraf) are only required during development.


Version Support

Only the latest stable release of Devflow receives security updates. Users are encouraged to stay current:

npm update -g @tjsasakinpm/devflow
Version Supported
latest ✅
older ❌

Response Time

This is an open-source project maintained by volunteers. We make best-effort attempts to:

  • Acknowledge receipt of a vulnerability report within 5 business days
  • Triage and provide an initial assessment within 10 business days
  • Release a fix within 30 days for confirmed HIGH or CRITICAL findings

No formal SLA applies. If you have not received a response within these timeframes, please follow up on the original report.

There aren't any published security advisories