Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

Appearance settings

Security: deploywhisper/deploywhisper

Security

SECURITY.md

DeployWhisper Security Policy

DeployWhisper analyzes infrastructure artifacts, deployment context, incident memory, scanner output, and AI-agent workflows. Treat all submitted artifacts and related context as potentially sensitive.

Reporting Vulnerabilities

Do not open public issues for vulnerabilities that could expose credentials, private infrastructure, unsafe parsing behavior, prompt-injection paths, or deployment-risk bypasses.

Use the most private available repository reporting path:

  1. GitHub private vulnerability reporting, if enabled for the repository.
  2. A private maintainer contact path listed by the repository owner or CODEOWNERS.
  3. If no private path is available, open a public issue with only a minimal, non-sensitive summary and ask maintainers to establish a private channel before sharing details.

Include enough information for maintainers to reproduce and assess the issue without sharing real secrets or production artifacts.

Supported Scope

Security reports may cover:

  • Secret handling and redaction failures.
  • Unsafe artifact persistence, logging, or prompt construction.
  • Parser behavior that mishandles untrusted input.
  • Prompt-injection or agent-output boundary failures.
  • Cross-project data leakage.
  • Authentication, authorization, or API exposure defects.
  • Supply-chain, release, or dependency concerns.

Local-First Boundary

DeployWhisper's default posture is self-hosted and local-first. Raw infrastructure artifacts should remain local by default. External model providers, connectors, or integrations must be explicit and should receive only the minimum safe context required.

Disclosure Expectations

Maintainers should acknowledge credible private reports, triage severity, and coordinate a fix before public disclosure. Public advisories should avoid exposing exploit details before users have a reasonable opportunity to update or mitigate.

Sensitive Data Guidance

When reporting an issue:

  • Use synthetic examples whenever possible.
  • Remove API keys, credentials, hostnames, customer names, incident details, and production identifiers.
  • Prefer minimal reproduction files over full real-world artifacts.
  • Call out whether the issue affects local-only mode, external provider mode, shared-team usage, or workflow integrations.

There aren't any published security advisories

Morty Proxy This is a proxified and sanitized view of the page, visit original site.