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 QBall-Inc/the-bulwark --skill issue-debugginggit clone --depth 1 https://github.com/QBall-Inc/the-bulwarkWrote 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/qball-inc/the-bulwark/issue-debugging)<a href="https://agentmods.dev/skills/qball-inc/the-bulwark/issue-debugging"><img src="https://agentmods.dev/badge/skills/qball-inc/the-bulwark/issue-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/qball-inc/the-bulwark/issue-debugging"><img src="https://agentmods.dev/badge/skills/qball-inc/the-bulwark/issue-debugging.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 313 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.02624 |
| Opus 5.5 | $0.00016 | $0.01050 |
| Sonnet 5.5 | $0.00008 | $0.00525 |
| Haiku 4.5 | $0.00004 | $0.00262 |
Grade A, and why
issue-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 — 391 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Issue Debugging
Systematic methodology for debugging that prevents the "first error fix and declare done" anti-pattern. Ensures root cause analysis, impact assessment, and verified fixes.
When to Use This Skill
Load this skill when the user request matches ANY of these patterns:
| Trigger Pattern | Example User Request |
|---|---|
| Bug investigation | "Why is X failing?", "Debug this error", "Investigate this issue" |
| Test failures | "Tests are failing", "Fix the broken tests", "Why did CI fail?" |
| Error analysis | "Getting this error...", "Exception in...", "Stack trace shows..." |
| Regression hunting | "This used to work", "Something broke after...", "Regression in..." |
| Fix validation | "Did the fix work?", "Verify this change", "Make sure it's fixed" |
| Root cause requests | "What's causing this?", "Find the root cause", "Why does this happen?" |
DO NOT use for:
- New feature implementation (no existing bug)
- Refactoring without reported issues
- General code questions
Overview
The Problem
AI (and developers) often fix the first visible error and declare victory without verifying end-user functionality works. This leads to:
- Symptom fixes instead of root cause fixes
- Regressions in seemingly unrelated areas
- False confidence in fix completeness
The Solution
This skill provides:
- Root Cause Analysis - Find the actual cause, not just symptoms
- Impact Analysis - Map affected areas and dependencies
- Validation Planning - Tiered test priorities (P1/P2/P3)
- Confidence Assessment - Objective fix reliability rating
- Escalation Triggers - When manual testing is required
Root Cause Analysis
The 5 Whys Method
Iteratively ask "Why?" to drill past symptoms to root cause:
Problem: Login fails with 500 error
├─ Why? → API returns null for user profile
├─ Why? → Database query found no records
├─ Why? → Migration didn't run on deploy
├─ Why? → Deployment script skipped migration step
└─ Why? → Environment variable missing
Root Cause: Missing environment variable (not the 500 error)
What ships with it
2 files 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.
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 · 391 lines · 39 tokens per session scan A f0250d5ffeea
issue-debugging is a skill published in the GitHub repository QBall-Inc/the-bulwark (8 stars, last pushed today), licensed MIT. It adds 39 tokens to every session and 2,624 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
ros2-troubleshooting
Diagnose ROS 2 Jazzy faults that reading the code cannot settle: QoS mismatch, sensor-mount and frame errors (REP 103/105), and odometry calibrated against CAD instead of the floor. Ships four runnable pass/fail checks.
concurrency-and-async
Use when a ticket involves work happening at the same time — async/await flows, promises and futures, goroutines and channels, threads and locks, parallel requests, races, deadlocks, "sometimes it double-processes" — and the change must be correct under interleaving, not just on the happy sequence. Invoke for any…
concurrency-review
Use as a REVIEW LENS when judging another agent's diff for concurrency and data-integrity defects under failure — concurrent writers, unsynchronised read-modify-write and lost updates, temp-file name collisions, non-atomic file writes, cross-process locking and stale-lock leases, crash/pause/retry schedules…
fix-bug
Use when a ticket reports defective behaviour — a wrong result, a crash, a regression, "X doesn't work when Y" — and the fix must be proven with a test that fails before and passes after. Invoke for any bug, incident follow-up, or "this used to work" ticket before touching the code.
structured-logging-and-tracing
Use when a ticket adds or changes what a service records about itself — log lines, metrics, spans, correlation ids — and the output must be structured, correlated across processes, safe (no secrets or personal data), and useful to the person debugging at 3am. Invoke for "add logging for X", "we can't see why Y fails"…
incident-response
Use when a production incident has been declared and needs classification, triage, escalation, and post-mortem. Covers SEV1-SEV4 severity, false-positive filtering, NIST SP 800-61 lifecycle, and blameless post-mortem facilitation. For proactive threat hunting before an incident fires, use threat-detection. For cloud…