Vetted · Skills
BrowseHow vetting worksPublish
Sign inGet started
Vetted · Skills

A marketplace for AI agent skills where every skill is security scanned before it can be published.

Marketplace

  • Browse skills
  • Publish a skill
  • Your dashboard

Trust

  • How vetting works
  • Open source (MIT)

Built as an open teaching project. Free to use and build on.

Automated vetting reduces risk. It is not a guarantee of safety.

Back to catalog
LegalVerified Safev1.1.0

NDA Redline Helper

A commercial lawyer's NDA playbook, so you know which clauses to push back on.

ARAisha Rahman· Commercial Counsel1,980 installsUpdated 2026-07-05

Security scan report

scanner@0.1.0

Verified Safe

Passed every automated check with no high-risk findings.

1 info
  • InfoInstruction patternSKILL.md

    Includes an appropriate non-advice disclaimer

    The skill correctly frames itself as drafting assistance and routes to counsel. No action needed.

Classifier: claude-haiku-4-5. Automated vetting reduces risk but is not a guarantee. Review a skill before trusting it with sensitive data.

SKILL.md

The full instructions that load into your agent. Read them before installing.


name: nda-redline-helper description: Review a mutual NDA against a standard playbook — flag one-sided terms, missing carve-outs, and unusual durations. Use when reviewing an inbound NDA.

NDA Redline Helper

Compare an inbound NDA to the playbook and produce a redline summary. This is drafting assistance, not legal advice — always route to counsel.

Flag

  • Definitions of Confidential Information that are overly broad.
  • Missing standard carve-outs (public info, independently developed, etc.).
  • Durations over 3 years for ordinary commercial information.
  • One-sided remedies or indemnities.

Cite the matching playbook clause for each flag.

Install

$npx vetted add @vetted/nda-redline-helper

The authenticated CLI arrives after beta. Downloads will require your account token.

Bundle contents

  • SKILL.md3.6 KB
  • playbook.md2.4 KB

Scripts are analyzed statically. We never execute a submitted skill.