Skip to content

Security: devsecblueprint/devsecops-claude-skill

Security

SECURITY.md

Security Policy

Reporting a vulnerability

Do not open a public issue.

Report privately through either:

Expect an acknowledgement within 5 business days.

What counts as a vulnerability here

This repository ships guidance, not a running service. The interesting failure modes are different from a typical application, and two of them matter more than anything else:

Unsafe generated configuration. A pipeline snippet in SKILL.md or examples/ that would introduce a real weakness if a user adopted it — a leaked secret pattern, an over-permissioned deployment identity, an unpinned action, a scanner whose exit code is discarded so findings cannot block. These are security defects in the product even though nothing here executes.

Guidance that neutralizes a control. A rule, example, or explanation that causes an agent to recommend disabling, bypassing, or silently downgrading a security control without an explicit, recorded exception.

Also in scope: prompt-injection paths that make the skill emit attacker-chosen pipeline configuration, and anything in tools/ that executes untrusted input or writes outside the repository.

Out of scope

  • Vulnerabilities in third-party products named as illustrative examples. Tool names in this repository are never requirements — report those upstream.
  • Disagreement with a rule's enforcement default. That is an engineering discussion: open a normal issue.
  • Findings in the DSB platform repository. Report those to devsecblueprint/devsecblueprint.

Disclosure

We will confirm the issue, agree a fix timeline with you, and credit you in the release notes unless you would rather stay anonymous. Please give us a reasonable window to ship a fix before publishing.

There aren't any published security advisories