Borrowing it
Nothing to install: this file belongs to webhook-co/webhook. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/webhook-co/webhook/main/.cursor/skills/systematic-debugging/SKILL.mdgit clone --depth 1 https://github.com/webhook-co/webhookWrote 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/webhook-co/webhook/systematic-debugging)<a href="https://agentmods.dev/skills/webhook-co/webhook/systematic-debugging"><img src="https://agentmods.dev/badge/skills/webhook-co/webhook/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/webhook-co/webhook/systematic-debugging"><img src="https://agentmods.dev/badge/skills/webhook-co/webhook/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 | $0.00056 | $0.00724 |
| Opus 5.5 | $0.00022 | $0.00290 |
| Sonnet 5.5 | $0.00011 | $0.00145 |
| Haiku 4.5 | $0.00006 | $0.00072 |
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 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.
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 — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Systematic debugging
Find the root cause before changing code. Guess-and-check edits hide bugs instead of fixing them and are especially dangerous on the correctness-critical paths (signing/verification, dedup, retries, ordering, replay, metering). Work the four phases in order.
Phase 1 — root-cause investigation
- Reproduce with the smallest case: endpoint setup, a redacted sample event, the exact request/command. Note environment and version.
- Follow the event id through ingest → DO → delivery using OpenTelemetry traces and logs. Inspect dedup/idempotency state, retry/backoff state, and signature verification.
- State what you observe vs what you expect. Don't theorize past the evidence yet.
Phase 2 — pattern analysis
- Is this a one-off or a class of bug? Search for the same shape elsewhere (other surfaces, other endpoints) — parity means a bug in one surface often exists in the others.
- Check recent changes and the blast radius. Distinguish retryable from terminal failure: a "bug" is sometimes correct retry behavior reacting to a real downstream failure.
Phase 3 — hypothesis testing
- Form one concrete, falsifiable hypothesis and design the cheapest test that would disprove it (a failing unit test, a targeted log/trace, a minimal repro).
- Change one variable at a time. Capture the result before forming the next hypothesis. Reproduce the bug reliably before attempting a fix — if you can't reproduce it, you can't confirm a fix.
Phase 4 — implementation
- Once the root cause is confirmed, fix it under TDD: write the regression test that fails without
the fix first (see
test-driven-development), then implement the minimal correct fix. - Verify the original repro is gone and nothing adjacent regressed.
The 3-attempt stop rule
If three genuine fix attempts have failed, stop thrashing. Repeated failed fixes mean the mental
model is wrong, not that the next tweak will work. Step back and do an architectural review: write
down what you know, what each attempt assumed, and which assumption keeps being wrong; reconsider the
design or boundary involved; and bring in a second perspective (e.g. the code-reviewer agent) before
trying again. More attempts on a broken model just dig the hole deeper.
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 · 60 lines · 56 tokens per session scan A 1fa90ef026c5
systematic-debugging is a skill published in the GitHub repository webhook-co/webhook (0 stars, last pushed 18d ago), licensed Apache-2.0. It adds 56 tokens to every session and 724 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
triage
Diagnose and fix broken MCP servers in mcptoon — doctor, health, per-server probes, and the common failure playbook.
metagit-workspace-grep
Search on-disk file contents across managed workspace repos via ripgrep (or Python fallback). Use when hunting code strings, configs, or Terraform/Docker patterns — not for manifest metadata.
metagit-upstream-scan
Use when a coding agent encounters likely upstream blockers and must find related workspace repositories, files, and probable root causes.
metagit-upstream-triage
Triage cross-repository blockers by ranking likely upstream repositories and files with metagit search and hinting tools. Use when local fixes appear incomplete.
scan
A source-code scanner that checks a specified folder, or the workspace root by default, using an MCP scanning tool. Its results are heuristic signals, not a legal conclusion.
api-errors
McpError constructor, JsonRpcErrorCode reference, and error handling patterns for @cyanheads/mcp-ts-core. Use when looking up error codes, understanding where errors should be thrown vs. caught, or using ErrorHandler.tryCatch in services.