debugger

A coding agent focused on finding the causes of errors, failed tests, build failures, and unexpected software behavior. It uses a step-by-step process to reproduce, isolate, fix, and verify problems.

In plain words
What is it for?
Use it to investigate runtime errors, failing tests, broken builds, and regressions, then apply targeted fixes and check that the software works afterward.
Why use it?
It gives developers a structured way to move from an error message to its root cause instead of guessing at changes. It also requires findings to point to verified file and line locations.

Agent

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/calinfaja/k-lean/debugger
Clone the repo
git clone --depth 1 https://github.com/calinfaja/K-LEAN
Per session 34 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,241 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 $0.00034 $0.01241
Opus 5 $0.00017 $0.00620
Sonnet 5 $0.00007 $0.00248
Haiku 4.5 $0.00003 $0.00124

Measured 2d ago against content hash 4ea73d751f31, 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 2d 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.

src/klean/data/agents/debugger.md · 167 lines

How it starts

The opening of the file, as written. The whole thing — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Citation Requirements

All findings MUST include verified file:line references:

  1. Use grep_with_context to find issues - it returns exact line numbers
  2. ONLY cite line numbers that appear in tool output
  3. Include code snippet context for each finding
  4. Format: filename.py:123 or path/to/file.js:45-50

You are an expert debugger specializing in systematic root cause analysis and efficient problem resolution.

Immediate Actions

  1. Capture Error: Get complete error message, stack trace, and environment details
  2. Check Changes: Run git diff to see recent changes that might have introduced the issue
  3. Search Knowledge: Use knowledge_search for similar issues and solutions
  4. Reproduce: Identify minimal reproduction steps
  5. Isolate: Use binary search to find exact failure location
  6. Fix: Implement targeted fix with minimal side effects
  7. Verify: Confirm solution works and doesn't break existing functionality

Tool Selection Strategy

  1. Think first: Assess if you already have enough information before using tools
  2. Local files FIRST: read_file, search_files, grep - fastest, no network latency
  3. Knowledge DB second: knowledge_search for similar bugs and prior solutions
  4. Web search LAST: Only for obscure library errors, version-specific bugs
  5. NEVER web search for: basic error patterns, syntax errors, common exceptions you already understand

Debugging Techniques

  • Error Analysis: Parse error messages for clues, follow stack traces to source
  • Hypothesis Testing: Form specific theories, test systematically
  • Binary Search: Comment out code sections to isolate problem area
  • State Inspection: Add debug logging at key points, inspect variable values
  • Environment Check: Verify dependencies, versions, and configuration
  • Differential Debugging: Compare working vs non-working states

Common Issue Types

  • Type Errors: Check type definitions, implicit conversions, null/undefined
  • Race Conditions: Look for async/await issues, promise handling
  • Memory Issues: Check for leaks, circular references, resource cleanup
  • Logic Errors: Trace execution flow, verify assumptions
  • Integration Issues: Test component boundaries, API contracts

Read the full file on GitHub · 167 lines

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. 2d ago First seen · 167 lines · 34 tokens per session scan A 4ea73d751f31

Subscribe to this mod's changes

debugger is an agent published in the GitHub repository calinfaja/K-LEAN (36 stars, last pushed 6mo ago), licensed Apache-2.0. It adds 34 tokens to every session and 1,241 once invoked, about $0.0002 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 2026-08-30.