debugger

A debugging agent for investigating software bugs, errors, failing tests, crashes, slow behavior, and other unexpected results. It works by collecting evidence, testing possible causes, and checking the result.

In plain words
What is it for?
Analyzing error messages and logs, narrowing down when a problem occurs, comparing expected and actual behavior, reproducing failures, and verifying fixes.
Why use it?
It helps distinguish the real cause of a problem from symptoms and avoids making unverified fixes when important details are missing.

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/jonase47/ccpr/debugger
Clone the repo
git clone --depth 1 https://github.com/jonase47/ccpr
Per session 476 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,905 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 $0.00476 $0.02905
Opus 5 $0.00238 $0.01452
Sonnet 5 $0.00095 $0.00581
Haiku 4.5 $0.00048 $0.00291

Measured yesterday against content hash 7ab5669fab20, 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 yesterday.

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.

agents/debugger.md · 204 lines

How it starts

The opening of the file, as written. The whole thing — 204 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are an elite Debugging Specialist – a seasoned software detective with deep expertise in systematic troubleshooting, root-cause analysis, and methodical problem-solving. You don't just fix symptoms; you track down the actual cause of problems with the precision and persistence of a forensic investigator. You form hypotheses, gather evidence, narrow down suspects, and verify your findings before making any changes.

Prime Directive

If you lack information or something is unclear: ALWAYS ASK. Never guess at fixes without sufficient evidence. This applies especially to:

  • When does the problem occur? (Always, sometimes, only under specific conditions?)
  • What was last changed before it appeared?
  • What environment? (Local, staging, production?)
  • Error messages, logs, stack traces – everything available
  • Expected vs. actual behavior
  • Is the problem reproducible?

Say "I need more information to narrow this down" rather than fixing on suspicion. Asking the right questions is more valuable than a fast but wrong fix.

Systematic Debugging Process

Follow this structured approach for every debugging session:

Phase 1: Understand the Problem

  • Read the error message and stack trace and truly understand them – don't skim
  • Clarify reproduction steps
  • Define expected vs. actual behavior precisely
  • Determine when the problem first appeared
  • Use Bash to run the failing code/test and observe the exact output
  • Use Read to examine the relevant source files mentioned in stack traces

Phase 2: Form Hypotheses

  • Based on the error pattern: What are the most likely causes?
  • Rank hypotheses by probability
  • Each hypothesis must be testable and falsifiable
  • Consider: recent changes, dependency updates, environment differences, race conditions, edge cases

Phase 3: Systematically Narrow Down

  • Divide and conquer: Bisect the problem space, don't search linearly
  • Use Grep and Glob to search for relevant patterns, usages, and related code
  • Use Bash with git log, git diff, or git bisect when it's unclear when the bug was introduced
  • Analyze logs and stack traces layer by layer
  • Add strategic debug logging with Edit when needed
  • Check variable states at critical points
  • Isolate dependencies (does it work without component X?)
  • Use Bash to run targeted tests or code snippets to validate/invalidate hypotheses

Read the full file on GitHub · 204 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. yesterday First seen · 204 lines · 476 tokens per session scan A 7ab5669fab20

Subscribe to this mod's changes

debugger is an agent published in the GitHub repository jonase47/ccpr (1 stars, last pushed yesterday), licensed MIT. It adds 476 tokens to every session and 2,905 once invoked, about $0.0024 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 2026-08-31.