Skills diagnosing-bugs
📦

diagnosing-bugs

Content revision r2 Safe 🌐 Network access⚙️ External commands

Diagnose Complex Bugs with a Reproducible Loop

Hard bugs waste time when investigation starts without a reliable reproduction. This skill builds a tight feedback loop, tests hypotheses, and verifies durable fixes.

Supports: Claude Codex Code(CC)
📊 75 Adequate

Install with my Agent

Copy this request to your Agent. It includes the canonical Skill page and manifest.

Agent request
Review the Skillstore skill "diagnosing-bugs" from https://skillstore.io/skills/mattpocock-diagnosing-bugs.md and its manifest at https://skillstore.io/api/skills/mattpocock-diagnosing-bugs/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.

Your Agent should still show its plan and request any confirmation required by the security policy.

Agent-readable resources

Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.

Markdown detail GET /skills/mattpocock-diagnosing-bugs.md Signed manifest GET /api/skills/mattpocock-diagnosing-bugs/manifest Signed lockfile GET /api/skills/mattpocock-diagnosing-bugs/lockfile

Test it

Using "diagnosing-bugs". A checkout action fails intermittently after a recent release.

Expected outcome:

  • Feedback loop: repeat the checkout scenario with fixed test data and record the exact payment error.
  • Minimized case: one discount code and one payment method reproduce the failure.
  • Leading cause: shared validation mutates the discount state before payment authorization.
  • Verification: the regression test passes, and the original checkout loop remains green across repeated runs.

Using "diagnosing-bugs". An API endpoint became slower after a dependency update.

Expected outcome:

  • Baseline: the repeatable request benchmark confirms a substantial latency increase.
  • Bisection: the first slow version introduces an additional remote lookup.
  • Fix: cache the stable lookup result within the request scope.
  • Verification: benchmark latency returns to baseline without changing response behavior.

Security Audit

Safe
v5 • 7/23/2026 Open versioned report

All ten static findings are false positives. The URL is a localhost example, shell-like tokens are Markdown code spans, and the Bash template only prompts for human input. No prompt injection, exfiltration intent, or other semantic security issue was found.

2
Files scanned
177
Lines analyzed
0
Review items
0
False positives ignored

Risk Factors

🌐 Network access (1)
scripts/hitl-loop.template.sh:31
⚙️ External commands (8)
SKILL.md:10 SKILL.md:27 SKILL.md:29 SKILL.md:58 SKILL.md:104 SKILL.md:106 SKILL.md:130 SKILL.md:134
No confirmed security findings were detected by the latest completed static and semantic audit. This does not prove the skill has no side effects.
Audited by: codex View Audit History →
Share & cite this report

Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.

Open versioned report
Security Assessment

Copy report link

https://skillstore.io/skills/mattpocock-diagnosing-bugs/audits/5?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_report

Markdown badge

[![Skillstore security assessment](https://skillstore.io/badges/skills/mattpocock-diagnosing-bugs/security.svg)](https://skillstore.io/skills/mattpocock-diagnosing-bugs?utm_source=security_passport_badge)

HTML badge

<a href="https://skillstore.io/skills/mattpocock-diagnosing-bugs?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/mattpocock-diagnosing-bugs/security.svg" alt="Skillstore security assessment" loading="lazy"></a>

Embed card

<iframe src="https://skillstore.io/embed/skills/mattpocock-diagnosing-bugs.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>
Academic citations (APA · BibTeX · CFF)

APA citation

mattpocock. (2026). diagnosing-bugs security audit report (audit version 5) [Author version unspecified]. Skillstore. https://skillstore.io/skills/mattpocock-diagnosing-bugs/audits/5

BibTeX citation

@techreport{mattpocock-mattpocock-diagnosing-bugs-2026, author = {mattpocock}, title = {diagnosing-bugs security audit report (audit version 5)}, institution = {Skillstore}, year = {2026}, number = {5}, url = {https://skillstore.io/skills/mattpocock-diagnosing-bugs/audits/5}, note = {Author version unspecified} }

CITATION.cff

cff-version: 1.2.0 message: "If you use this Skill, cite its author and this versioned security audit report." title: "diagnosing-bugs security audit report (audit version 5)" version: "unspecified" type: report authors: - name: "mattpocock" date-released: "2026-07-23" url: "https://skillstore.io/skills/mattpocock-diagnosing-bugs/audits/5" identifiers: - type: other value: "skillstore:mattpocock-diagnosing-bugs:audit:5" description: "Skillstore immutable audit report identifier"

Skillstore Score

Why this score Evidence Confidence: Medium
45
Architecture
85
Maintainability
87
Content
67
Community
83
Spec Compliance

What You Can Build

Debug Application Failures

Reproduce a difficult application failure, minimize the scenario, rank causes, and verify a focused fix.

Trace Performance Regressions

Establish a performance baseline, bisect changes, and validate that the regression is removed.

Create Durable Regression Coverage

Turn a reported defect into a deterministic regression test at the correct system boundary.

Try These Prompts

Reproduce a Bug 
Diagnose this bug: [symptom]. First create and run the smallest reliable pass-or-fail loop. Show the command, result, and exact assertion.
Minimize and Rank Causes 
Use this failing reproduction: [command and output]. Minimize it one variable at a time. Then present three to five ranked, falsifiable hypotheses with predicted outcomes.
Analyze a Performance Regression 
Diagnose this performance regression: [baseline and current behavior]. Build a repeatable benchmark, measure a baseline, bisect likely changes, and test one variable at a time.
Complete Diagnosis and Prevention 
Investigate [complex failure] using a deterministic feedback loop. Instrument only boundaries that separate ranked hypotheses. Produce the root cause, regression test, verified fix, cleanup evidence, and one prevention recommendation.

Best Practices

  • Create and run a red-capable feedback loop before forming detailed theories.
  • Change one variable at a time and connect every probe to a falsifiable prediction.
  • Re-run both the regression test and the original scenario before declaring completion.

Avoid

  • Do not read large areas of code before establishing a reliable reproduction.
  • Do not collect broad logs without a hypothesis that explains what each probe distinguishes.
  • Do not accept a passing shallow test when it does not exercise the real failure pattern.

Frequently Asked Questions

What does this skill diagnose?
It covers functional bugs, intermittent failures, incorrect output, thrown errors, and performance regressions.
Can it apply a fix automatically?
It guides diagnosis and fix verification. Available tools and repository permissions determine which actions the agent can perform.
Does it support intermittent bugs?
Yes. It raises the reproduction rate through repetition, controlled stress, pinned randomness, and narrower timing conditions.
How does it handle performance problems?
It establishes a repeatable baseline, uses profiling or query plans, bisects changes, and measures the result after each fix.
Does it require an existing test suite?
No. It can use a failing test, command-line harness, browser script, captured trace, or structured human procedure.
How does it determine that work is complete?
The original reproduction no longer fails, regression coverage passes, temporary instrumentation is removed, and the root cause is documented.

Developer Details

Author

mattpocock

License

MIT

Skillstore revision

r2

Version notice

The author did not declare a version.

Repository

https://github.com/mattpocock/skills/tree/main/skills/engineering/diagnosing-bugs/

Ref

ebdfe608f5de2b66ff37ab4af12af8ac4f5e8006

Maintenance freshness

7/25/2026

Usage

7 downloads · 0 views

File structure

📁 scripts/

📄 hitl-loop.template.sh

📄 SKILL.md