speckit-red-team-run

Run adversarial red-team reviews on functional specifications with parallel lens-driven agents.

Updated Mar 16, 2026
One-click install
npx skills add https://github.com/B0yZ4kr14/OrthoPlus-Enterprise --skill speckit-red-team-run
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: speckit-red-team-run
Source: https://github.com/B0yZ4kr14/OrthoPlus-Enterprise/tree/main/.specify-backups/20260518-141826/_agents/skills/speckit-red-team-run
Command: npx skills add https://github.com/B0yZ4kr14/OrthoPlus-Enterprise --skill speckit-red-team-run

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps teams uncover hidden weaknesses in functional specifications by running an adversarial “red team” review before architecture decisions get locked in, then organizing the results into actionable resolution paths.

Core Features & Use Cases

  • Parallel lens-based attack: Runs multiple adversarial lenses selected from a configurable lens catalog to cover different risk categories.
  • Spec-driven triggering & selection: Automatically picks lenses by heuristically matching triggers in the target spec, or lets you explicitly override lenses for controlled runs.
  • Structured findings & resolution workflow: Writes a findings report under the feature’s directory and guides maintainers through classifying resolutions (spec-fix, new open question, accepted risk, or out-of-scope) without auto-editing historical SpecKit working records.

Quick Start

Run a red-team review on a target functional spec by executing: /speckit.red-team.run specs/<feature-id>/spec.md --yes

Frequently Asked Questions about speckit-red-team-run

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I run an adversarial red-team review on a functional spec?

To run an adversarial red-team review on a functional spec, execute the command targeting your spec file path. The review uses parallel, lens-driven agents to uncover hidden spec weaknesses and aggregates structured findings for maintainer resolution.

What is a red-team spec review and how does it work?

A red-team spec review adversarially tests functional specifications before architecture lock-in. It applies multiple configurable risk lenses in parallel to identify vulnerabilities and outputs a structured findings report classifying resolutions.

Can I override the automatic lens selection for a spec review?

Yes, you can explicitly override lenses for controlled red-team runs. By default, the system automatically selects lenses by heuristically matching triggers in the target spec, but explicit overrides bypass this automatic trigger matching.

Does the red-team review automatically fix the functional specification?

No, the red-team review does not auto-edit historical SpecKit working records. It writes a deterministic findings report to the feature directory and guides maintainers through classifying resolutions like spec-fix, new open question, or accepted risk.

Can I use this adversarial spec review in CI pipelines?

Yes, you can use this adversarial spec review in CI pipelines through its dry-run reporting mode. It writes deterministic reports to the feature directory and enforces per-lens finding bounds, supporting both CI and interactive workflows.

What do I need to set up before running a SpecKit red-team review?

You need a SpecKit project with a functional spec target and a configurable red-team lens catalog YAML. The system locates the target spec, parses the lens catalog, and performs trigger matching to execute the review.