codedb CLAUDE.md

A set of repository instructions for codedb, a Zig-based code-intelligence server. It describes the project layout, test commands, test style, and rules for filing issues.

In plain words
What is it for?
Use it when developing codedb, running its Zig tests, or preparing a bug report with a reproducible failing test.
Why use it?
Code bugs are easier to reproduce and fix when every issue includes a failing test and follows the project’s workflow. These instructions make that process explicit.

Instructions file

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/justrach/codedb/claude-md
Clone the repo
git clone --depth 1 https://github.com/justrach/codedb
Per session 412 This file is loaded in full into every session.
When invoked 412 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 $0.00412 $0.00412
Opus 5 $0.00206 $0.00206
Sonnet 5 $0.00082 $0.00082
Haiku 4.5 $0.00041 $0.00041

Measured 2d ago against content hash 36b4b39771eb, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

codedb CLAUDE.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 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.

CLAUDE.md · 39 lines

What it actually says

codedb — Agent Instructions

Project

Zig 0.16.x code intelligence server. Tests live in the split src/test_*.zig files — one binary per area (see build.zig). Build and test with zig build test; run a single binary with e.g. zig build test-index.

Rules

Filing Issues

Every GitHub issue must include a failing test case. No exceptions.

When creating an issue:

  1. Write a test "issue-XX: <description>" block in the matching src/test_*.zig file (e.g. src/test_index.zig for index bugs) that fails on the current main branch
  2. Verify it fails: zig build test 2>&1 | grep "issue-XX"
  3. File the issue via gh issue create with this structure:
    • Title: <module>: <concise description>
    • Body sections: Problem, Failing Test (the zig test block), Expected, Fix
    • Labels: bug for defects, priority:p0 for crashes, priority:p2 for correctness
  4. Commit the failing test on a branch: issue-XX-failing-test
  5. Do not fix the bug in the same commit as the failing test

If you cannot write a failing test, the issue is not well-defined enough to file.

Test Style

  • Use std.testing and testing.allocator
  • Use std.heap.ArenaAllocator for Explorer tests
  • Always defer cleanup (arena.deinit, allocator.free)
  • One test per issue, named test "issue-XX: <short description>"
  • Keep tests minimal — only exercise the specific broken code path

Code Style

  • No comments or documentation changes unless explicitly asked
  • Prefer minimal, targeted fixes over refactors
  • Follow existing patterns in the module you're editing
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 · 39 lines · 412 tokens per session scan A 36b4b39771eb

Subscribe to this mod's changes

codedb CLAUDE.md is an instructions file published in the GitHub repository justrach/codedb (1,370 stars, last pushed yesterday), licensed BSD-3-Clause. It adds 412 tokens to every session, about $0.0021 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.