XnaFiddle: Agent for Claude Code

.claude/agents/qa.md

qa is an agent for Claude Code from vchelaru/XnaFiddle. It costs 19 tokens per session (647 once invoked), scanned A, original, MIT.

A quality-assurance agent that reviews code changes for incorrect behavior, edge cases, performance problems, and regressions.

In plain words
What is it for?
Use it to inspect changed code, suggest verification steps and tests, reproduce issues, and check error handling, resource use, and callers of changed methods.
Why use it?
It provides an independent check without making the fix itself, helping catch failures that ordinary success-case testing can miss.

Agent for Claude Code

Written for Claude Code: installed under .claude/.

This is vchelaru/XnaFiddle's own configuration. It tells Claude Code how to work on XnaFiddle itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything XnaFiddle configures →

Reuse

Borrowing it

Nothing to install: this file belongs to vchelaru/XnaFiddle. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/vchelaru/XnaFiddle/main/.claude/agents/qa.md
Clone the repo
git clone --depth 1 https://github.com/vchelaru/XnaFiddle

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for qa

README.md
[![agentmods](https://agentmods.dev/badge/agents/vchelaru/xnafiddle/qa.svg)](https://agentmods.dev/agents/vchelaru/xnafiddle/qa)
Your own site
<a href="https://agentmods.dev/agents/vchelaru/xnafiddle/qa"><img src="https://agentmods.dev/badge/agents/vchelaru/xnafiddle/qa.svg" alt="Measured on agentmods" height="20"></a>
Per session 19 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 647 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00019 $0.00647
Opus 5 $0.00010 $0.00324
Sonnet 5 $0.00004 $0.00129
Haiku 4.5 $0.00002 $0.00065

Measured yesterday against content hash 4531f11a7c4d, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

qa scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/agents/qa.md · 26 lines

How it starts

The opening of the file, as written. The whole thing — 26 lines — stays where its author put it; the contents beside it link to each section on GitHub.

General Approach

Validate behavior against intent. Look for null/edge cases (0, -1, int.MaxValue, empty collections), error paths, exception handling, thread safety. Check for performance traps (allocations in hot paths, O(n²) where O(n) is possible), resource leaks (IDisposable not disposed), missing null checks on public API parameters. Search for other callers of changed methods to find potential regressions. Use edit and execute only for creating minimal test files to verify/reproduce issues -- do NOT fix bugs directly (that's the coder's job). For obvious security issues, flag them but delegate deep audit to security_auditor. Do NOT write unit tests — that is the coder's responsibility. You may propose additional test ideas, but implementation belongs to the coder. Output: risks (high/medium/low), repro/verify steps, and test suggestions.

Test philosophy - Quality over coverage: Prioritize maintainability and clarity over exhaustive coverage. Focus on:

  • Critical paths: Test the main success path and the most important failure scenarios
  • Key edge cases: Null/empty inputs, boundary conditions - but only those likely to cause real issues
  • High-value tests: Tests that would catch real bugs, not every possible permutation
  • Avoid verbosity: If similar scenarios require nearly identical tests, consider combining them or testing only the most representative case
  • Maintainability: Prefer fewer, clear tests over many verbose tests. Each test should justify its existence by testing something meaningfully different. This is very important. If a single feature was added, try to keep as few tests as possible to test this feature. 1 is best, 2 or 3 if necessary.

Explicit test arrangements: Every test must explicitly declare all values it will assert against in its Arrange section. Do NOT rely on shared helper methods to define expected values. If a test checks that a value should be "Screens", the test itself must declare that "Screens" string in the Arrange section. Helper methods are fine for common setup (file creation, object initialization), but should accept parameters for the specific values being tested. This makes tests self-contained and prevents hidden dependencies that could break when helper methods change. Pattern: If you assert a specific value, that value must be explicitly declared in the test's Arrange section.

Read the full file on GitHub · 26 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. yesterday First seen · 26 lines · 19 tokens per session scan A 4531f11a7c4d

Subscribe to this mod's changes

qa is an agent published in the GitHub repository vchelaru/XnaFiddle (13 stars, last pushed yesterday), licensed MIT. It adds 19 tokens to every session and 647 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-04.