Now liveThe Skillselion MCP - thousands of ranked skills, loaded into your agent mid-task. No install.Get it →
Platform Engineering avatar

Sup Standards

  • Updated June 11, 2026
  • mholtzhausen/claude-market-mhsup

sup-standards is a Claude Code skill in the Security category. Org-wide engineering standards: Conventional Commit and secure-code-review skills, a /pr-description command, a code-reviewer subagent, and a PreToolUse secret-blocking hook.

Key points

  • sup-standards
  • Security
  • AI-coding skill

Sup Standards by the numbers

  • Data as of Jul 7, 2026 (Skillselion catalog sync)
/plugin marketplace add mholtzhausen/claude-market-mhsup
/plugin install sup-standards@sup-internal

Add your badge

Show developers this skill is listed on Skillselion. Paste this into your README.

Listed on Skillselion
Last updatedJune 11, 2026
Repositorymholtzhausen/claude-market-mhsup

What it does

Org-wide engineering standards: Conventional Commit and secure-code-review skills, a /pr-description command, a code-reviewer subagent, and a PreToolUse secret-blocking hook.

README.md

sup-standards

Org-wide engineering standards for Superbalist, packaged as a Claude Code plugin. Maintained by Platform Engineering.

Version: 0.1.0

What's inside

Type Name What it does
Skill conventional-commit Teaches Claude to write Conventional Commits 1.0.0 messages — correct type(scope): subject, imperative subjects, wrapped bodies, BREAKING CHANGE footers. Triggers when you draft or reword a commit.
Skill secure-coding-review A framework-agnostic security checklist (input validation, authn/authz, secrets, injection, unsafe deserialization, dependency risk). Triggers when reviewing code for security.
Command /pr-description Reads the branch/staged diff and produces a PR description with Summary, Changes, Testing, and Risk sections.
Agent code-reviewer A read-only subagent that reviews a diff for correctness, security, and standards adherence, returning findings grouped by severity. It never modifies files.
Hook block-secrets A PreToolUse hook that scans Write/Edit/MultiEdit content and blocks the change if it contains obvious secrets.

The secret-blocking hook

hooks/block-secrets.sh runs before every file write or edit. It scans the proposed content for high-signal secret patterns:

  • AWS access key IDs (AKIA…) and secret-key assignments
  • Private key headers (-----BEGIN … PRIVATE KEY-----)
  • GitHub and Slack tokens
  • High-entropy values assigned to variables named secret, token, password, or api_key

On a match it blocks the edit and explains what matched. It is non-destructive — it only reads the payload and prints a message; it never deletes or rewrites anything.

Overriding the hook

If you hit a genuine false positive (for example, a documented test fixture), set the escape-hatch environment variable for that session:

export ALLOW_SECRET=1

To disable the hook entirely, disable the plugin (or remove the hooks block); prefer the scoped ALLOW_SECRET=1 override instead.

Updating this plugin

Every change must bump version in .claude-plugin/plugin.json (semver) and add an entry to CHANGELOG.md. See the repo-root CONTRIBUTING.md for the full workflow.

Related skills

Securityappsec

This week in AI coding

Five minutes, every Monday - the tools, releases and tactics for developers.

unsubscribe anytime.