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 bendrucker/claude --skill diagnosing-bugsgit clone --depth 1 https://github.com/bendrucker/claudeWrote 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/bendrucker/claude/diagnosing-bugs)<a href="https://agentmods.dev/skills/bendrucker/claude/diagnosing-bugs"><img src="https://agentmods.dev/badge/skills/bendrucker/claude/diagnosing-bugs/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/bendrucker/claude/diagnosing-bugs"><img src="https://agentmods.dev/badge/skills/bendrucker/claude/diagnosing-bugs.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 | $0.00055 | $0.00432 |
| Opus 5.5 | $0.00022 | $0.00173 |
| Sonnet 5.5 | $0.00011 | $0.00086 |
| Haiku 4.5 | $0.00006 | $0.00043 |
Grade A, and why
diagnosing-bugs 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.
What it actually says
Diagnosing Bugs
Diagnose the failure in $ARGUMENTS, or the one the user just reported.
Reproduce
Build one command that fails on this bug. Reading and grepping to work out how to invoke the failing path is how you build it, not a detour around it.
A command that shows only that nothing crashed does not count. It has to drive the real code path and produce the symptom that was reported.
For a failing CI or pipeline job, that command already ran. Use github:actions-monitor or gitlab:ci-monitor to get the logs rather than reconstructing the job locally.
When nothing you try reaches the bug, say so and stop. Name what you tried and what each attempt could not reach, then ask for an environment where it reproduces, a captured artifact, or permission to add instrumentation.
Find the Cause
Change one thing per run and let the command settle it. A cause the command has not responded to is a guess, and saying which line is at fault is not the same as showing the command go green when you change it.
Report one cause. When the cause is not clear, say what you ruled out and how you ruled it out, then ask rather than presenting a list to choose from.
Hand Off
With the cause established and the fix clear, enter plan mode and map out the implementation.
When the user asked you to diagnose without changing anything, stop at the cause and the plan.
Confirm
Once a fix is in, run the original reproduction again. Report it fixed only after that command passes, and quote the run.
A fix that was never run against the thing that failed is not a fix, and neither is one verified against a smaller case you built along the way.
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 · 38 lines · 55 tokens per session scan A b450d90b388d
diagnosing-bugs is a skill published in the GitHub repository bendrucker/claude (18 stars, last pushed today), licensed MIT. It adds 55 tokens to every session and 432 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 2026-09-03.
Other skills, from other repositories
tidewave-integration
Tidewave MCP runtime tools — debugging, smoke testing, live state inspection, SQL queries, hex docs. Use when evaluating code in a running Phoenix app.
bug-fixer
A bug reproduces, the expected behaviour is clear, and the fix stands apart from the current work.
release-it
Build production-ready systems with stability patterns: circuit breakers, bulkheads, timeouts, and retry logic. Use when the user mentions "production outage", "circuit breaker", "deployment pipeline", "chaos engineering", "retry storm", "health checks", "my service keeps crashing", "prevent cascading failures", or…
debug
Interactive debugging workflow with hypothesis-driven probe loop. Use when: unknown bugs, script errors, silent failures, troubleshooting. Not for: known bugs (use bug-fix), GitHub issue analysis (use issue-analyze), code understanding (use code-explore). Output: debug report with probe journal + root cause + fix.
git-investigate
Git history investigation. Use when: tracking code changes, finding where bugs were introduced, root cause analysis. Not for: code exploration (use code-explore), issue analysis (use issue-analyze). Output: history trace + root cause report.
simplify
Wrap-up refactoring — simplify code, eliminate duplication, preserve behavior.