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.
npx skills add MadAppGang/claude-code --skill systematic-debugginggit clone --depth 1 https://github.com/MadAppGang/claude-codeWrote 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.
[](https://agentmods.dev/skills/madappgang/claude-code/systematic-debugging)<a href="https://agentmods.dev/skills/madappgang/claude-code/systematic-debugging"><img src="https://agentmods.dev/badge/skills/madappgang/claude-code/systematic-debugging/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/madappgang/claude-code/systematic-debugging"><img src="https://agentmods.dev/badge/skills/madappgang/claude-code/systematic-debugging.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00039 | $0.04489 |
| Opus 5.5 | $0.00016 | $0.01796 |
| Sonnet 5.5 | $0.00008 | $0.00898 |
| Haiku 4.5 | $0.00004 | $0.00449 |
Grade A, and why
systematic-debugging scanned grade A with 1 finding 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 1mo 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
response = requests.get('https://api.example.com/users') How it starts
The opening of the file, as written. The whole thing — 577 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Systematic Debugging
Iron Law: "NO FIXES WITHOUT ROOT CAUSE INVESTIGATION FIRST"
When to Use
Use this skill when:
- A test fails and you need to understand why
- An error is thrown and you need to find the cause
- A feature behaves unexpectedly
- Performance degrades and you need to identify bottlenecks
- Data corruption occurs and you need to trace the source
- A bug reappears after "fixing" it
Red Flags (Violation Indicators)
Detect these patterns that indicate skipping root cause investigation:
- Fix without understanding - "I'll just add a null check" (why is it null?)
- Skip to solution - "Let me try wrapping this in setTimeout" (why does timing matter?)
- Restart tools - "Let me restart the dev server" (what state is corrupted?)
- Clear cache - "Let me clear the cache" (what cache entry is stale?)
- Change multiple things - "Let me update these 3 files" (which one fixes it?)
- Shouldn't cause problem - "This change shouldn't affect that" (but it does, why?)
- Assume cause - "Must be a race condition" (what evidence supports this?)
Key Concepts
1. Root Cause vs. Symptom
Symptom: What you observe (test fails, error thrown, wrong output) Root Cause: Why it happens (null value, wrong condition, missing await)
Example:
Symptom: "TypeError: Cannot read property 'name' of undefined"
Root Cause: API returns null when user not found, but code expects object
Bad approach: Add user?.name (fixes symptom, not cause)
Good approach: Add validation if (!user) throw new NotFoundError() (fixes cause)
2. Data Flow Tracing
Principle: Follow data from source to error point
Steps:
- Identify error location (stack trace line number)
- Identify data involved (variable name, object property)
- Trace backwards: Where does this data come from?
- Find divergence: Where does actual differ from expected?
Example:
Error: "Expected 'active' but got 'inactive'"
Location: user.test.ts:42 - expect(user.status).toBe('active')
Data: user.status = 'inactive'
Trace: user.status ← updateUser() ← API response ← database
Divergence: Database has status='inactive' (expected 'active')
Root Cause: Test setup didn't create user with active status
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.
- 1mo ago First seen · 577 lines · 39 tokens per session scan A 174f089baeb7
systematic-debugging is a skill published in the GitHub repository MadAppGang/claude-code (286 stars, last pushed 7mo ago), licensed MIT. It adds 39 tokens to every session and 4,489 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.
Other skills, from other repositories
debug-agent
Systematic root cause analysis for bugs, errors, and unexpected behavior. Use when user reports something broken, an error they can't explain, or unexpected output.
fault-localizer
A fault-finding tool that uses an error stack or failed tests to narrow the problem to likely root-cause code. It uses test-failure patterns and the structure of the source code to navigate relevant areas.
cli-anything-nslogger
CLI harness for NSLogger — parse, filter, export, and monitor NSLogger log files (.rawnsloggerdata / .nsloggerdata).
cli-anything-nsight-graphics
Windows-first CLI harness for Nsight Graphics capture, GPU Trace summary, and ngfx-replay analysis.
trace
Use when encountering bugs, test failures, runtime errors, broken builds, or "this doesn't work" reports. Systematic root-cause analysis before any patch — never blind-patches symptoms. Standalone, ends with a final-integration review of the fix. Trigger with /hyperflow:trace, "debug this", "find the root cause", "why…
analyze-logs
Analyze application logs from the .evlog/logs/ directory. Use when debugging errors, investigating slow requests, understanding request patterns, or answering questions about application behavior. Reads structured NDJSON wide events written by evlog's file system drain.