adversarial-qa SKILL.md skillA

adversarial-qa from romabeckman/harness-kit is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add romabeckman/harness-kit/adversarial-qa@git:20260903.e393457

Pins git:20260903.e393457 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
romabeckman/harness-kit
Path
skills/adversarial-qa/SKILL.md
Versions recorded
5
Latest version
git:20260903.e393457, committed Sep 3, 2026
First version recorded
Aug 4, 2026
Size
9,023 bytes
Repository stars
30
License
MIT

Autonomous Adversarial QA agent. Reads machine-readable specs and code to execute edge-case and security testing, returning a JSON verdict.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

You are the **Adversarial QA Engineer**. Your goal is to break the implementation by finding edge cases, boundary faults, and security vulnerabilities (e.g., injections, race conditions, unhandled nulls) that standard TDD missed.

## EXECUTION MODE SWITCH
Before executing, detect how you were invoked:
1. **Autonomous Mode (Default when called by autonomous-orchestrator):** Read `${featureId}`, `${domain}`, `${projectPaths}`, and **`${scoreThresholdAdv}`** from the runtime context injection passed by the orchestrator. Use `${domain}` to locate spec documents at `docs/specs/${domain}/`. Set `featureId` in JSON output to `${featureId}`. Skip all interactive prompts.
2. **Interactive Mode:** Used ONLY when invoked directly by a human. Ask for the domain/feature context if not provided.

---

## SCORE THRESHOLD CONTEXT (Dynamic Validation Gate)
**In Autonomous Mode**, your `score` output will be compared against `${scoreThresholdAdv}` (injected by autonomous-orchestrator during Phase C):
- **`score >= ${scoreThresholdAdv}`** with no HIGH/CRITICAL vulnerability → This validator passes; the orchestrator still evaluates the technical-lead result
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add romabeckman/harness-kit/adversarial-qa@sha256:fe0d99599432b12e

The line at the top of this page pins the label git:20260903.e393457, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_oelf7yysdbi6omvc.svg)](https://markdownregistry.com/a/art_oelf7yysdbi6omvc)

4 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 5 versions of this file, between Aug 4, 2026 and Sep 3, 2026, about one every 7 days.

versioncommittedcommitsizeaudit
git:20260903.e393457 latest2026-09-03 e393457 9,023 BA view · diff
git:20260827.102c0592026-08-27 102c059 8,916 BA view · diff
git:20260810.b9eb2342026-08-10 b9eb234 7,788 BA view · diff
git:20260804.2d833b82026-08-04 2d833b8 7,525 BA view · diff
git:20260804.b7724222026-08-04 b772422 5,363 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (9023 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

romabeckman/harness-kit · 30 stars · license MIT · pushed 2026-11-07 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_oelf7yysdbi6omvc
GET https://markdownregistry.com/api/v1/resolve?ref=romabeckman/harness-kit/adversarial-qa
GET https://markdownregistry.com/api/v1/blob/fe0d99599432b12e6189cd10ecc532cd6bde12c507c5c24a618a722405315bf1

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from romabeckman/harness-kit

AGENTS.md@sdk AGENTS.md file
romabeckman/harness-kit · sdk/AGENTS.md
git:20260927.e804b61 · audit A · 30 stars
autonomous-orchestrator SKILL.md skill
romabeckman/harness-kit · skills/autonomous-orchestrator/SKILL.md · Sovereign loop manager. Handles file initialization, feature lifecycle tracking, and recursive…
git:20260901.fd66ff0 · audit A · 30 stars
harness-evaluator SKILL.md skill
romabeckman/harness-kit · skills/harness-evaluator/SKILL.md · Harness performance evaluator. Reads all execution traces in docs/harness-history/traces/, computes composite scores…
git:20260810.b9eb234 · audit A · 30 stars
harness-tracer SKILL.md skill
romabeckman/harness-kit · skills/harness-tracer/SKILL.md · Execution trace recorder. Captures what happened during a skill session and persists structured logs to…
git:20260810.b9eb234 · audit A · 30 stars
memory-context SKILL.md skill
romabeckman/harness-kit · skills/memory-context/SKILL.md · Initialize and maintain a local multi-project Harness Memory cache, then use Harness Memory MCP efficiently for…
git:20260930.84c4dd7 · audit A · 30 stars
meta-harness SKILL.md skill
romabeckman/harness-kit · skills/meta-harness/SKILL.md · Autonomous Meta-Harness proposer. Reads the full harness history filesystem, diagnoses failure patterns, proposes a…
git:20260810.b9eb234 · audit A · 30 stars
pbb-design SKILL.md skill
romabeckman/harness-kit · skills/pbb-design/SKILL.md · Designs a Product Backlog using the Product Backlog Building (PBB) methodology from an initial product or feature scope.
git:20260918.5fe8074 · audit A · 30 stars
project-memory SKILL.md skill
romabeckman/harness-kit · skills/project-memory/SKILL.md · Orchestrate sequential creation, updates, and evidence-backed consolidation of project documentation in docs/ and…
git:20261107.ac8b580 · audit A · 30 stars
qa-orchestrator SKILL.md skill
romabeckman/harness-kit · skills/qa-orchestrator/SKILL.md · Plan and run evidence-backed acceptance tests with the Harness Kit `hrns qa` CLI. Use when the user wants to install or…
git:20260920.f1e08a0 · audit B · 30 stars
read-ui-prototype SKILL.md skill
romabeckman/harness-kit · skills/read-ui-prototype/SKILL.md · UI/UX prototype reader. Analyzes interface prototypes (screens, Figma links, images, frames) and translates them into a…
git:20260810.3b77f15 · audit A · 30 stars
scope-refinement SKILL.md skill
romabeckman/harness-kit · skills/scope-refinement/SKILL.md · DDD Scope Refinement Orchestrator — coordinates all Domain-Driven Design phases from business discovery to test…
git:20260919.96d8340 · audit A · 30 stars
tdd-orchestrator SKILL.md skill
romabeckman/harness-kit · skills/tdd-orchestrator/SKILL.md · Orchestrates development workflow using Test-Driven Development (TDD) methodology. Coordinates between testing and…
git:20260919.96d8340 · audit A · 30 stars

Every file in romabeckman/harness-kit

Other files named adversarial-qa

adversarial-qa SKILL.md skill
cunhaax/ai-workflow · plugins/ai-workflow/skills/adversarial-qa/SKILL.md · Exploratory QA of a running feature through its UI (Playwright) and/or API (curl) to find issues the plan and committed…
git:20261004.cccdee6 · audit A · 5 stars

Browse by kind, by grade A, or by owner.