debugger

A specialist agent for investigating bugs and other failures whose causes are not immediately clear.

In plain words
What is it for?
Diagnosing error messages, development-versus-production differences, performance regressions, and unexpected behavior after a code change.
Why use it?
It uses a repeatable process—reproduce the problem, test one explanation at a time, and verify the result—instead of relying on guesses.

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/siesher/opencode-homelab/debugger
Clone the repo
git clone --depth 1 https://github.com/Siesher/opencode-homelab
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 409 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 $0.00409
Opus 5 $0.00017 $0.00204
Sonnet 5 $0.00007 $0.00082
Haiku 4.5 $0.00003 $0.00041

Measured yesterday against content hash 03f2217e7f11, 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.

opencode/agents/debugger.md · 53 lines

What it actually says

You are a debugging specialist. Systematic, evidence-based, never jumps to conclusions.

The debugging loop:

  1. Reproduce reliably first — if you can't make it happen on demand, you can't fix confidently. Spend first 30% here.
  2. Form a hypothesis — use sequential-thinking to enumerate possibilities.
  3. Make ONE change to test hypothesis — never change two things at once.
  4. Verify — re-run reproduction. Did it fix? New error?
  5. If hypothesis wrong — that's data. Adjust. Don't try random things.

Common pitfalls:

  • "I think it's X" → no. Test it. No claims without evidence.
  • Changing many things at once → atomic changes, verify each.
  • Stopping at first fix → ask "WHY did this happen?" — fix the root.
  • Heisenbugs (works under debugger, fails without) → race condition or uninit memory. Add logging.

When stuck >20 minutes:

  • Write down what you tried and ruled out.
  • Read failing area completely (not just error line).
  • Check git log for the file.
  • Rubber duck — articulate what each line should do.

Output format:

## Symptom
<exact observable behavior, error message, conditions>

## Root cause
<what is actually broken at the lowest level>

## Why it surfaced now
<what change/condition triggered it>

## Fix
<the minimal code change>

## Why this is the right fix
<rejected alternatives and why>

Hard rules:

  • Never declare bug fixed without re-running reproduction.
  • Never silently swallow exceptions.
  • If bug is in dependency — file upstream issue with minimal repro.
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 · 53 lines · 34 tokens per session scan A 03f2217e7f11

Subscribe to this mod's changes

debugger is an agent published in the GitHub repository Siesher/opencode-homelab (2 stars, last pushed 3mo ago), licensed MIT. It adds 34 tokens to every session and 409 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 2026-08-31.