quantum-debugger

A methodical guide for finding difficult software bugs, including intermittent crashes, race conditions, memory leaks, deadlocks, and slowdowns. It forms hypotheses, tests them, isolates the cause, and documents the result.

In plain words
What is it for?
Use it to investigate crashes, concurrency problems, memory errors, performance regressions, and the code change that introduced a failure, with tools such as gdb, Valgrind, perf, sanitizers, tracing tools, and git bisect.
Why use it?
It replaces guesswork with measured experiments, which is useful when a bug disappears during observation or cannot be reproduced reliably.

Skill for Claude CodeCodex

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/mturac/hermes-supercode-skills/quantum-debugger
Any agent
npx skills add mturac/hermes-supercode-skills --skill quantum-debugger
Clone the repo
git clone --depth 1 https://github.com/mturac/hermes-supercode-skills

Made for: Claude Code, Codex.

Per session 143 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,619 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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 $0.00143 $0.01619
Opus 5 $0.00072 $0.00809
Sonnet 5 $0.00029 $0.00324
Haiku 4.5 $0.00014 $0.00162

Measured 2d ago against content hash e2731e62d563, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

quantum-debugger 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 2d ago.

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.

skills/quantum-debugger/SKILL.md · 189 lines

How it starts

The opening of the file, as written. The whole thing — 189 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Quantum Debugger

You are a debugging specialist. You approach every bug with the scientific method: observe, hypothesize, test, isolate, fix, and document. You never guess at root causes — you generate hypotheses, design experiments to test them, and follow the evidence.

The name "quantum" reflects the nature of the hardest bugs: they change behavior when you observe them (heisenbugs), they exist in superposition (intermittent failures), and they require precise measurement to collapse into a definitive root cause.

Debugging Tools Reference

Know when to reach for each:

Tool Use for
gdb / lldb Source-level debugging, core dump analysis
valgrind --tool=memcheck Memory leaks, use-after-free, buffer overflows
valgrind --tool=helgrind Race conditions, lock ordering violations
perf record + perf report CPU profiling, flamegraph generation
strace / ltrace System call and library call tracing
AddressSanitizer (-fsanitize=address) Memory errors at compile time
ThreadSanitizer (-fsanitize=thread) Data races at compile time
git bisect Finding the commit that introduced a bug
bpftrace / eBPF Dynamic kernel and userspace tracing

For interpreted languages (Python, Node.js, Ruby), the equivalents are language-specific profilers and debuggers — but the methodology is the same.

Workflow — The Scientific Debugging Method

1. Observe

Gather all available evidence before forming any hypotheses:

  • Error messages — exact text, not paraphrased
  • Stack traces — full trace, not just the top frame
  • Logs — surrounding context, not just the error line
  • Timing — when does it happen? Under what load? How often?
  • Environment — OS, language version, dependencies, hardware
  • Reproducer — can you trigger it reliably? If not, what's the closest?

Ask the user for anything missing from this list. The quality of the investigation depends entirely on the quality of the initial evidence.

Read the full file on GitHub · 189 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. 2d ago First seen · 189 lines · 143 tokens per session scan A e2731e62d563

Subscribe to this mod's changes

quantum-debugger is a skill published in the GitHub repository mturac/hermes-supercode-skills (2 stars, last pushed 3mo ago), licensed MIT. It adds 143 tokens to every session and 1,619 once invoked, about $0.0007 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 2026-08-31.

Related

Other skills, from other repositories

kaspa-sovereign-architect-engine

Source-grounded Kaspa protocol and Toccata engineering for Rusty Kaspa, KIPs, wallets, Kasware/Kaspium, indexers, WASM/TypeScript, covenants, ZK, vProgs, mining, infrastructure, repository audits, and current network or release status. Use when Codex must research, explain, design, implement, review, or verify Kaspa…

gryszzz/Kaspa-Ai-Agent-Skill · 114 tokens

swift-networking

Builds a protocol-based async/await networking layer with URLSession, testable abstractions, error handling, and mock support for Swift projects. Use when user says "add networking", "create an API layer", "fetch data from API", "build a network service", "add URLSession", "implement HTTP requests", "create a REST…

jeremieb/swift-unit-test-instructions · 82 tokens

swift-architecture-audit

Audits an existing Swift/SwiftUI/UIKit codebase for MVVM compliance, testability issues, architectural violations, concurrency problems, and anti-patterns. Use when user says "audit the codebase", "review the architecture", "why is this hard to test", "onboard to this project", "analyze the code structure", "find…

jeremieb/swift-unit-test-instructions · 88 tokens

swift-code-review

Reviews Swift, SwiftUI, and UIKit code for MVVM compliance, testing standards, naming conventions, async correctness, and anti-patterns. Produces structured feedback grouped by severity. Use when user says "review this code", "check this PR", "code review", "is this correct", "give me feedback on", "review my…

jeremieb/swift-unit-test-instructions · 83 tokens

swift-testing

Writes unit tests and UI tests for Swift, SwiftUI, and UIKit projects following enterprise or indie testing standards. Use when user says "write tests", "add unit tests", "test this code", "create test cases", "add XCTest", "write Swift Testing tests", "help me test this", or "add tests for". Applies enterprise or…

jeremieb/swift-unit-test-instructions · 87 tokens

swiftui-component

Creates SwiftUI views and screens following MVVM with thin view layers, ViewModel extraction, previews, accessibility support, and loading/error states. Use when user says "create a SwiftUI view", "build a screen", "add a new view", "create a component", "build the UI for", "add a SwiftUI screen", or "make a new…

jeremieb/swift-unit-test-instructions · 81 tokens