systematic-debugging is agent-read markdown (skill) from jnpiyush/agentx: Disciplined debugging for stalled bug fixes. Use when an Engineer or Reviewer has tried two or more fixes that did not work, when symptoms keep moving, or when the same test keeps failing for different reasons. Forces root-cause investigation, pattern analysis, hypothesis testing, and a final implementation, in that order..
Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.
What the file says
# Systematic Debugging
> WHEN: A bug fix has failed two or more times, the same test keeps failing for different reasons, or the symptom keeps moving when fixes are applied. Also use proactively for any `type:bug` that touches more than one file or crosses a module boundary.
## When to Use This Skill
Load this skill when:
- Two or more fixes for the same bug have not held
- The failing test now fails in a different way after each attempt
- A `type:bug` issue is non-trivial (more than a single-file typo)
- A Reviewer has issued `CHANGES REQUESTED` twice on the same defect
- You feel the urge to "try one more thing" instead of investigating
Skip when:
- The bug is a single-line obvious typo with no dependencies
- The failure mode is fully understood and the fix is mechanical
## Prerequisites
- Reproduction of the bug, or a failing test that captures it
- Access to the relevant logs, stack traces, and recent commits
- Time budget; this skill is slower than guess-and-fix, on purpose
## Rationalization Table
The three-failed-fixes rule exists because LLMs and humans both reach for these instead of investigating.
| Rationalization | Reality |
|-----------------|---------|
…
Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.
GET https://markdownregistry.com/api/v1/artifacts/art_w4zhtfw7jbx7y5dm
GET https://markdownregistry.com/api/v1/resolve?ref=jnpiyush/agentx/systematic-debugging
GET https://markdownregistry.com/api/v1/blob/e2b8a8bf69f6ec2037916be054c5fb73f0dccfadb6a1490ba0fd1ab66859efc9
Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
jnpiyush/agentx · .github/skills/ai-systems/agent-observability/SKILL.md · Instrument LLM agents with tracing, metrics, and evaluation telemetry. Use when adding OpenTelemetry GenAI semantic…
jnpiyush/agentx · .github/skills/ai-systems/ai-agent-development/SKILL.md · Build production-ready AI agents with Microsoft Foundry and Agent Framework. Use when creating AI agents, selecting LLM…
jnpiyush/agentx · .github/skills/ai-systems/ai-evaluation/SKILL.md · Use when evaluating AI quality, safety, completeness and cost with held-out tasks, executable graders, calibrated…
jnpiyush/agentx · .github/skills/ai-systems/ai-safety-and-red-teaming/SKILL.md · Defend LLM systems against prompt injection, jailbreaks, data exfiltration, and unsafe output. Covers input/output…
a5c-ai/babysitter · library/methodologies/superpowers/skills/systematic-debugging/SKILL.md · Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes. Requires root cause…
