debug-mode

A hypothesis-driven workflow for investigating difficult runtime bugs with targeted logging and collected evidence. Runtime bugs are problems that appear while a program is running, such as timing issues or unexpected state changes.

In plain words
What is it for?
Use it when investigating race conditions, asynchronous behaviour, corrupted state, or other bugs that require reproducing the problem and examining runtime logs.
Why use it?
It replaces guesses with tested explanations when a problem is intermittent, hard to reproduce, or not clear from reading the code alone.

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/nettee/ai-collection/debug-mode
Any agent
npx skills add nettee/ai-collection --skill debug-mode
Clone the repo
git clone --depth 1 https://github.com/nettee/ai-collection

Made for: Claude Code, Codex.

Per session 52 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,210 The whole file, excluding the scripts and references it only reads on demand.
Security scan D 4 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 $0.00052 $0.02210
Opus 5 $0.00026 $0.01105
Sonnet 5 $0.00010 $0.00442
Haiku 4.5 $0.00005 $0.00221

Measured 2d ago against content hash fb9d75f3be64, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade D, and why

debug-mode scanned grade D with 4 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.

The scan reads SKILL.md. This mod also ships 1 executable file (server.js), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Sends data to an external URLmediumData exfiltration

A POST to an outside endpoint may be telemetry or may be exfiltration; either way the mod talks to somewhere, and you should know where.

fetch("http://localhost:9742/debug", { method: "POST", headers: { "Content-Type": "application/json" }, body: JSON.stringify({ hypothesisId: "H1", location: "processOrder:entry", message: "checking initial state", data:

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

mkdir -p .debug-mode && cp "$(find ~/.claude/plugins -path '*/debug-mode/server.js' -print -quit 2>/dev/null || find .claude-plugin -path '*/debug-mode/server.js' -print -quit 2>/dev/null || find plugin/skills -path '*/d

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

rm -rf .debug-mode/

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -s http://localhost:9742/health
content/skills/debug-mode/SKILL.md · 281 lines

How it starts

The opening of the file, as written. The whole thing — 281 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Debug Mode: Hypothesis-Driven Runtime Debugging

Overview

Debug Mode is a structured debugging workflow inspired by Cursor's Debug Mode. Instead of guessing at fixes, you form hypotheses about root causes, instrument code with targeted logging, collect runtime evidence, then make minimal, precise fixes.

Core principle: Evidence first, fix second.

When to Use

  • Runtime bugs that are hard to reproduce or understand from code alone
  • Race conditions, async/timing issues
  • State corruption or unexpected values at runtime
  • Bugs where the root cause isn't obvious from reading code
  • Intermittent failures that need runtime data to diagnose

Methodology

The workflow has 7 steps:

  1. Understand the bug — Read the bug description and relevant code
  2. Generate hypotheses — Form 3-5 specific theories about the root cause, labeled H1-H5
  3. Instrument code — Add fetch() calls at strategic locations, each tagged with which hypothesis it tests
  4. Start debug server — Copy and run the server that collects logs to .debug-mode/debug.log
  5. Reproduce — Ask user to reproduce the bug while instrumentation collects evidence
  6. Evaluate hypotheses — Read logs, mark each hypothesis CONFIRMED or REJECTED with evidence
  7. Fix and clean up — Apply minimal fix for confirmed hypothesis, remove all instrumentation

If the fix doesn't work, loop back to step 3 with additional hypotheses or more instrumentation.


Workflow

Step 1: Create Task List

Use TodoWrite to create a task list:

  • Understand bug and read codebase
  • Generate hypotheses
  • Start debug server and instrument code
  • Ask user to reproduce
  • Analyze logs and evaluate hypotheses
  • Propose and verify fix
  • Remove all instrumentation and cleanup

Step 2: Understand the Bug

If the bug description is vague or unclear, ask the user:

  • What is the unexpected behavior?
  • What steps reproduce it?
  • What is the expected behavior?
  • What environment / runtime?

Read relevant source files to understand the suspected code paths. Use Grep and Glob to find related code.

Read the full file on GitHub · 281 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

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 · 281 lines · 52 tokens per session scan D fb9d75f3be64

Subscribe to this mod's changes

debug-mode is a skill published in the GitHub repository nettee/ai-collection (24 stars, last pushed 1mo ago), licensed MIT. It adds 52 tokens to every session and 2,210 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it D with 4 findings (sends data to an external url, reads agent configuration directories, recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Related

Other skills, from other repositories

systematic-debugging

Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.

obra/superpowers · 21 tokens

chat-pet-sprite-creation

Use when creating or changing VS Code chat pet sprite art, sprite sheets, state animations, eye treatments, Stable/Insiders variants, or pet transitions under src/vs/workbench/contrib/chat/browser/widget/media/chatPet.

microsoft/vscode · 53 tokens

cpu-profile-analysis

Analyze V8/Chrome CPU profiles (.cpuprofile) and DevTools trace files (Trace-.json). Use when: profiling performance, investigating slow functions, comparing code paths, finding bottlenecks, analyzing timeToRequest, understanding call trees from sampling profiler data, analyzing layout/paint/rendering, investigating…

microsoft/vscode · 71 tokens

next-cache-components-adoption

Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…

vercel/next.js · 95 tokens

babysit-pr

Babysit a GitHub pull request after creation by continuously polling review comments, CI checks/workflow runs, and mergeability state until the PR is merged/closed or user help is required. Diagnose failures, retry likely flaky failures up to 3 times, auto-fix/push branch-related issues when appropriate, and keep…

openai/codex · 114 tokens

imagegen

Generate or edit raster images when the task benefits from AI-created bitmap visuals such as photos, illustrations, textures, sprites, mockups, or transparent-background cutouts. Use when Codex should create a brand-new image, transform an existing image, or derive visual variants from references, and the output…

openai/codex · 113 tokens