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 clubpay/ronykit --skill systematic-debugginggit clone --depth 1 https://github.com/clubpay/ronykitWrote 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/clubpay/ronykit/systematic-debugging)<a href="https://agentmods.dev/skills/clubpay/ronykit/systematic-debugging"><img src="https://agentmods.dev/badge/skills/clubpay/ronykit/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/clubpay/ronykit/systematic-debugging"><img src="https://agentmods.dev/badge/skills/clubpay/ronykit/systematic-debugging.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00068 | $0.00678 |
| Opus 5.5 | $0.00027 | $0.00271 |
| Sonnet 5.5 | $0.00014 | $0.00136 |
| Haiku 4.5 | $0.00007 | $0.00068 |
Grade A, and why
systematic-debugging 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 23d 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.
How it starts
The opening of the file, as written. The whole thing — 74 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Systematic Debugging
Guessing creates new bugs and hides old ones. Find the root cause first.
The iron rule
No fix without a root-cause hypothesis you have confirmed with evidence.
If you cannot explain why the bug happens, you are not ready to fix it.
When to use
Any test failure, crash, wrong output, flaky test, performance regression, build failure, or integration issue — especially when under time pressure or when a previous fix didn't hold.
Phase 1 — Understand
- Read the full error and stack trace; note file, line, and exact message.
- Reproduce it reliably. A bug you can't reproduce, you can't verify fixed.
- Establish the smallest reproduction and the expected vs actual behavior.
Phase 2 — Investigate
- Form a hypothesis about the cause and predict what you'd observe if true.
- Gather evidence: targeted logging, a debugger,
git bisect, binary search by commenting/disabling, or diffing a working vs broken state. - Follow the data flow backward from the symptom to its origin. Question assumptions ("this can't be nil" — prove it).
Phase 3 — Fix the cause
- Change the root cause, not the symptom. Patching where it crashed instead of where the bad value originated just moves the bug.
- Make the smallest change that addresses the cause.
Phase 4 — Verify & prevent
- Write a test that fails before the fix and passes after.
- Re-run the full relevant suite to check for regressions.
- Ask whether the same class of bug exists elsewhere.
Tactics
git bisectto locate the introducing commit.- Binary search the input/code path to isolate the trigger.
- Add structured logging at decision points; remove it before committing.
- For flakiness: suspect time, ordering, concurrency, and shared state; run with
-raceand-count=20 -run '^TestName$'. - In a RonyKit workspace, check the cheap mechanical causes first: stale sqlc
code (
make sqlc), stale stubs after a contract change (make gen-stub), missing settings/env forx/settings, and fx wiring errors printed at startup (a missing provider names the type it could not build). - Use
x/telemetrytraces and logkit fields (request/trace IDs) to follow a request across services instead of adding ad-hoc prints.
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.
- 23d ago Changed · +10 lines · +27 tokens per session 959a5d709226
- 1mo ago First seen · 64 lines · 41 tokens per session scan A 1f5380c6e817
systematic-debugging is a skill published in the GitHub repository clubpay/ronykit (38 stars, last pushed today), licensed BSD-3-Clause. It adds 68 tokens to every session and 678 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-08-30.
Other skills, from other repositories
config-loader-helper
Diagnose configuration-related failures, enumerate required env vars, and guide safe local test setup (no secrets).
fix-suggester
Diagnose failures and propose minimal, test-backed fixes with verification and rollback instructions.
pinchtab-dev
Develop and contribute to the PinchTab project. Use when working on PinchTab source code, adding features, fixing bugs, running tests, or preparing PRs. Triggers on "work on pinchtab", "pinchtab development", "contribute to pinchtab", "fix pinchtab bug", "add pinchtab feature".
debugging-failures
Diagnose a failing Ginkgo suite as an agent — always run with --json-report into a predictable temp/gitignored location, read the terminal verdict line, then use jq to extract structured failure details (name, message, file:line, panic value, captured logs). Covers the panicked-vs-failed trap, panic locations pointing…
dev-cycle
Crawlberg iteration loops codified as Taskfile tasks — alef install/generate/format/bump, core and binding builds, e2e generate/build/test cycles, cleanup tiers, and the mock-server / stale-.so / precompiled-NIF / generated-e2e gotchas. Load when running or debugging crawlberg build, alef regeneration, or e2e…
ci-orchestrator
Run a CI-like pipeline locally (format, lint, vet, static-analysis, tests) and summarize per-step results with remediation guidance.