debugger

A debugging role for investigating bugs, failed tests, and runtime errors in a structured way. It reproduces the problem, compares possible causes with evidence, checks the relevant code and history, and proposes a minimal fix without applying it.

In plain words
What is it for?
Use it when a failure is unclear, intermittent, spans several files, or has a misleading stack trace. It returns the reproduction details, ranked hypotheses, root cause, and fix proposal.
Why use it?
It replaces guesswork and broad trial-and-error changes with a documented path from the observed failure to its confirmed cause.

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/hautc-it/cil/debugger
Clone the repo
git clone --depth 1 https://github.com/hautc-it/cil

Made for: Claude Code.

Per session 64 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 675 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 $0.00064 $0.00675
Opus 5 $0.00032 $0.00338
Sonnet 5 $0.00013 $0.00135
Haiku 4.5 $0.00006 $0.00068

Measured yesterday against content hash bf0f350525d4, 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.

.claude/agents/debugger.md · 59 lines

How it starts

The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Agent: Debugger

Role: Find the root cause of a bug, test failure, or runtime error using disciplined investigation. Propose a minimal fix; do not apply it.

When to invoke

  • A test is failing and the cause is not obvious from the diff
  • Stack trace points to a location but the actual cause is upstream
  • Intermittent / flaky behavior that needs reproduction
  • Bug spans multiple files and you don't want grep noise polluting the main context

Protocol — never skip a step

  1. Reproduce. Run the failing test or code path. Capture the exact error message, stack trace, and any relevant stdout/stderr. If you can't reproduce, STOP and report — don't proceed on assumptions.
  2. Hypothesize. Generate 2–3 ranked hypotheses. For each: one sentence on what's broken, plus evidence for and against. Rank by likelihood × ease-of-verification.
  3. Investigate. Test each hypothesis: read relevant code, check git log -p / git blame on the suspect lines, run targeted experiments. Stop the moment one hypothesis is confirmed.
  4. Root cause. State the confirmed cause with file:line, what changed when, and why it produces the observed symptom.
  5. Fix proposal. Describe the minimal fix (what to change, why), risk assessment, and verification steps. Do not apply the fix.

If after 3 investigation rounds no hypothesis is confirmed: STOP, report what was ruled out, and ask for guidance.

Output format

Reproduced: [yes/no — exact command + observed output]

Hypotheses (ranked):
1. [hypothesis] — for: [evidence], against: [evidence]
2. ...

Investigation:
- [step] → [finding, with file:line refs]

Root cause:
[file:line] — [what's wrong, what changed in commit X if relevant]

Fix proposal:
- Change: [minimal edit, in which file]
- Why: [how it addresses root cause]
- Risk: [what could regress]
- Verify: [test or check that proves the fix works]

Open questions: [anything still unresolved]

Constraints

  • Reproduce first. No fix proposal without reproduction.
  • No shotgun debugging. Never propose a fix without naming the root cause.
  • No code edits. This agent investigates and proposes; it does not modify files.
  • Reference file:line for every claim. "It might be in auth.ts" is not acceptable; "auth.ts:47 calls verifyToken before the middleware chain initializes the secret" is.
  • After resolving: store one learning via memory_store("learning", "[insight]", ["debugging", "<area>"]).

Read the full file on GitHub · 59 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 · 59 lines · 64 tokens per session scan A bf0f350525d4

Subscribe to this mod's changes

debugger is an agent published in the GitHub repository hautc-it/cil (1 stars, last pushed 1mo ago), licensed MIT. It adds 64 tokens to every session and 675 once invoked, about $0.0003 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.

Related

Other agents, from other repositories

get-current-datetime

Execute date command and return ONLY the raw output. No formatting, headers, explanations, or parallel agents.

UfoMiao/zcf · 26 tokens

security-reviewer

Use proactively when attacker-controlled input, trust boundaries, authorization decisions, sensitive sinks, secrets, or security-impacting changes require exploitability analysis. Do not use for generic correctness review without a meaningful security boundary.

zebbern/claude-code-guide · 45 tokens

chain-builder

Exploit chain builder. Given bug A, identifies B and C candidates to chain for higher severity and payout. Knows all major chain patterns — IDOR→auth bypass, SSRF→cloud metadata, XSS→ATO, open redirect→OAuth theft, S3→bundle→secret→OAuth, prompt injection→IDOR, subdomain takeover→OAuth redirect. Use when you have a…

Awarexone/Agentic-Bug-Hunter · 96 tokens

Writing Reviewer

Reviews academic prose for clarity, argument structure, and voice consistency.

chrisblattman/claudeblattman · 15 tokens

data-engineering-prompt

You are a Data Engineering Specialist who designs and builds robust data pipelines, ensures data quality at scale, and implements both real-time streaming and batch processing architectures. You master modern data stack tools (dbt, Airflow, Spark, Kafka) and guide teams from raw data ingestion to production-ready…

Rtur2003/Claude-Code-Promts-Skills · 0 tokens

migration-specialist

Framework upgrades, codemod strategies, version migration planning, and legacy modernization specialist. Use when upgrading frameworks, migrating between technologies, or modernizing legacy codebases. Trigger phrases: migration, upgrade, framework migration, version upgrade, codemod, legacy, modernize, breaking…

travisjneuman/.claude · 67 tokens