debug-bank AGENTS.md

debug-bank AGENTS.md is an instructions file for Codex, OpenCode from soleimanmansouri/debug-bank. It costs 548 tokens per session, scanned A, original, MIT.

A shared debugging protocol for coding assistants. It defines seven stages: check known patterns, reproduce the bug, form and test explanations, isolate the cause, fix it, and record what was learned.

In plain words
What is it for?
Use it before investigating bugs in a repository with pattern and example folders, to guide diagnosis and document the final root cause.
Why use it?
It gives bug investigations a repeatable path and a stop rule when repeated fixes make no progress.

Instructions file for CodexOpenCode

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 instructions/soleimanmansouri/debug-bank/agents-md
Clone the repo
git clone --depth 1 https://github.com/soleimanmansouri/debug-bank

Made for: Codex, OpenCode.

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 debug-bank AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/soleimanmansouri/debug-bank/agents-md.svg)](https://agentmods.dev/instructions/soleimanmansouri/debug-bank/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/soleimanmansouri/debug-bank/agents-md"><img src="https://agentmods.dev/badge/instructions/soleimanmansouri/debug-bank/agents-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 548 This file is loaded in full into every session.
When invoked 548 The same file — it is already loaded in full.
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 $0.00548 $0.00548
Opus 5 $0.00274 $0.00274
Sonnet 5 $0.00110 $0.00110
Haiku 4.5 $0.00055 $0.00055

Measured 3d ago against content hash 68e22e52c5af, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

debug-bank AGENTS.md 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 3d 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.

AGENTS.md · 70 lines

How it starts

The opening of the file, as written. The whole thing — 70 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Debug Bank — Cross-Agent Instructions

This file provides the debug trajectory protocol for any AI coding agent (Codex CLI, Gemini CLI, Cursor, Windsurf, etc.).

Core Protocol

Before Investigating Any Bug

  1. Read files in patterns/ directory for matching root cause patterns (P01-P21+)
  2. Search examples/ for similar symptoms in domain catalogs
  3. Only then start: Reproduce → Hypothesize → Isolate → Diagnose → Fix → Record

The 7-Step Trajectory

Step Action Output
1. Pattern Check Scan pattern bank for matching root cause Match or no-match (30 seconds)
2. Reproduce Get exact error with full output Logs, stack trace, HTTP status
3. Hypothesize 2-3 ranked, falsifiable root causes Ordered list with test for each
4. Isolate Test hypotheses one at a time Narrowed to single area
5. Diagnose Identify single root cause, trace call chain Root cause statement
6. Fix Minimal change addressing root cause Code diff
7. Record Add trajectory to domain catalog Structured entry

Stop Rule

If 3 rounds of iterative fixing show no progress, STOP. Re-plan from scratch or switch strategy.

Recording Format

After fixing, add to the relevant domain catalog:

### [Category] Short Title (YYYY-MM-DD)
- **Symptom:** What was observed
- **Root cause:** The actual technical cause
- **Fix:** What was changed
- **Key insight:** The generalizable lesson
- **Pattern:** P-number if matches existing pattern

Feedback Rules

When corrected by a user, create a persistent rule:

Rule statement.

**Why:** The reason this matters.
**How to apply:** When/where to apply this rule.

Agent-Specific Notes

Codex CLI

  • Store patterns in .codex/patterns/ or project root patterns/
  • Reference this file as AGENTS.md in your project root

Gemini CLI

  • Store patterns in .gemini/patterns/ or project root patterns/
  • Gemini CLI reads AGENTS.md automatically

Read the full file on GitHub · 70 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. 3d ago First seen · 70 lines · 548 tokens per session scan A 68e22e52c5af

Subscribe to this mod's changes

debug-bank AGENTS.md is an instructions file published in the GitHub repository soleimanmansouri/debug-bank (1 stars, last pushed 3mo ago), licensed MIT. It adds 548 tokens to every session, about $0.0027 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.