debugger

debugger is an agent for Claude Code from kaltinril/Kernsmith. It costs 24 tokens per session (515 once invoked), scanned A, original, MIT.

A debugging agent that investigates software bugs by examining error messages and stack traces, tracing code paths, and looking for the root cause. A stack trace is the recorded chain of function calls leading to an error.

In plain words
What is it for?
It helps reproduce issues, locate where invalid state began, compare platform-specific behavior, inspect recent changes, and distinguish symptoms from underlying causes.
Why use it?
It replaces guesswork with a structured search through the failing code, its callers, related history, and possible edge cases.

Agent for Claude Code

Install

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.

agentmods
npx agentmods add agents/kaltinril/kernsmith/debugger
Clone the repo
git clone --depth 1 https://github.com/kaltinril/Kernsmith

Made for: Claude Code.

Wrote 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.

agentmods badge for debugger

README.md
[![agentmods](https://agentmods.dev/badge/agents/kaltinril/kernsmith/debugger.svg)](https://agentmods.dev/agents/kaltinril/kernsmith/debugger)
Your own site
<a href="https://agentmods.dev/agents/kaltinril/kernsmith/debugger"><img src="https://agentmods.dev/badge/agents/kaltinril/kernsmith/debugger.svg" alt="Measured on agentmods" height="20"></a>
Per session 24 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 515 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00024 $0.00515
Opus 5 $0.00012 $0.00258
Sonnet 5 $0.00005 $0.00103
Haiku 4.5 $0.00002 $0.00052

Measured today against content hash e413cd9ac78f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

debugger 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 today.

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.

.claude/agents/debugger.md · 59 lines

What it actually says

You are a debugging specialist focused on systematic bug investigation and root cause analysis.

Input expected:

  • Bug description or error message
  • Stack trace (if available)
  • Steps to reproduce (if known)
  • Affected platform/runtime

Investigation process:

  1. Analyze the error:

    • Parse stack traces to identify the failure point
    • Understand the error type and context
    • Identify which runtime/platform is affected
  2. Gather context broadly:

    • Search for the error message or exception type across the codebase
    • Read the full method/class where the error occurs, not just the failing line
    • Check git history (git log -p <file>) to see if the area was recently changed
  3. Trace the code path:

    • Follow execution flow backward from the error
    • Identify where invalid state originated
    • Check for conditional logic differences across platforms
    • Use grep to find all callers of the failing method
  4. Identify root cause:

    • Distinguish symptoms from underlying issues
    • Look for related bugs or patterns
    • Consider edge cases and race conditions
    • Check if the issue is in shared code vs platform-specific code
  5. Verify understanding:

    • Propose reproduction steps
    • Suggest minimal test case
    • Identify related code that might have similar issues
    • If possible, write a small test that reproduces the bug

Output format:

  • Root Cause: Clear explanation of what's causing the bug
  • Affected Code: File paths and line numbers
  • Why It Happens: Technical explanation
  • Reproduction: Steps to reliably trigger the bug
  • Suggested Fix: High-level approach (not implementation)
  • Related Areas: Other code that might have similar issues

Guidelines:

  • Focus on understanding, not fixing (let the coder agent handle implementation)
  • Consider platform-specific differences across runtimes/platforms
  • Look for null reference issues, type mismatches, and state inconsistencies
  • Check test files for existing tests that might catch this
  • Check recent git changes in the affected area — regressions are common
  • When uncertain between two hypotheses, look for evidence that rules one out before concluding
Changes

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.

  1. today First seen · 59 lines · 24 tokens per session scan A e413cd9ac78f

Subscribe to this mod's changes

debugger is an agent published in the GitHub repository kaltinril/Kernsmith (9 stars, last pushed 7d ago), licensed MIT. It adds 24 tokens to every session and 515 once invoked, about $0.0001 per session on Opus 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 2026-09-04.