memorax-code AGENTS.md

memorax-code AGENTS.md is an instructions file for Codex, OpenCode from memorax-ai/memorax-code. It costs 1,657 tokens per session, scanned A, original, MIT.

Repository-specific working rules that guide a coding agent through safe changes, architecture decisions, hooks, sessions, scope, data, and user-facing boundaries.

In plain words
What is it for?
Use it when working on the MemoraX Code repository, especially before changing architecture, agent behavior, documentation, or lifecycle and migration code.
Why use it?
It reduces accidental edits, ignored local changes, unsafe tests, and conflicts between current code and outdated generated notes.

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/memorax-ai/memorax-code/agents-md
Clone the repo
git clone --depth 1 https://github.com/memorax-ai/memorax-code

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 memorax-code AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/memorax-ai/memorax-code/agents-md.svg)](https://agentmods.dev/instructions/memorax-ai/memorax-code/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/memorax-ai/memorax-code/agents-md"><img src="https://agentmods.dev/badge/instructions/memorax-ai/memorax-code/agents-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 1,657 This file is loaded in full into every session.
When invoked 1,657 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 $0.01657 $0.01657
Opus 5 $0.00829 $0.00829
Sonnet 5 $0.00331 $0.00331
Haiku 4.5 $0.00166 $0.00166

Measured today against content hash db74ac93c306, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

memorax-code 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 today.

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 · 141 lines

How it starts

The opening of the file, as written. The whole thing — 141 lines — stays where its author put it; the contents beside it link to each section on GitHub.

MemoraX Code Agent Guide

This file defines the durable working rules for coding agents in this repository. Current user instructions, live source, and executable tests take precedence over generated memory or historical context.

1. Start Safely

  • Run git status --short --branch before editing. Confirm the intended worktree and preserve unrelated tracked, untracked, and staged changes.
  • Read the nearest implementation and tests before changing behavior. Prefer a small, layered change with an explicit verification target.
  • Do not mix unrelated formatting, renaming, dependency updates, or lockfile rewrites into the task. Do not run broad auto-fix commands unless required.
  • Use English for source identifiers, comments, docstrings, and canonical public documentation. Keep README.md and README.zh.md synchronized for user-facing changes.
  • Treat .repo_memory/ as local retrieval guidance, not current-code authority. It is Git-ignored and must not leak into public artifacts.
  • Use isolated MEMORAX_CODE_HOME, CODEX_HOME, CLAUDE_CONFIG_DIR, and OPENCODE_CONFIG_DIR locations for lifecycle, install, migration, or destructive tests.

2. Architecture Routing

ARCHITECTURE.md is the canonical map for system shape, package ownership, runtime flows, Backend source and test placement, dependency direction, stable root surfaces, and executable architecture contracts.

Read it before adding or moving modules or tests, changing entrypoints, control/data flow, state or authority, packaging/materialization, or cross-package or cross-capability dependencies.

Preserve the documented boundaries. When an intentional architecture change is required, update ARCHITECTURE.md and the affected executable contract in the same change, then run the matching verification profile in Section 5.

3. Hook, Session, and Scope Invariants

  • Hook commands are versioned and client-qualified. Required session, turn, or prompt correlation must be validated at the HTTP boundary; incomplete, conflicting, unknown, or client-inapplicable identities fail closed.
  • Codex rollout JSONL, Claude Code and CodeBuddy/WorkBuddy transcript JSONL, DSH's exact persisted Session Event Log interval, matching OpenCode SDK session-message records, and Trae's validated UserPromptSubmit/Stop Hook pair are the content authorities for their respective automatic writeback. Hook or plugin text is not a fallback outside Trae's narrow primary-authority exception; local trace, latest-turn guesses, and another client's format are never fallbacks.
  • Session, turn metadata, trace, and operational identity always include the client. Equal native IDs from different clients must remain isolated.
  • A live session is pinned to its resolved workspace and repository scope. Linked worktrees of one repository may share repository scope; unrelated repositories and genuine non-Git workspaces keep separate local identity. The documented direct-.git exception may use local-folder scope when its internal metadata is malformed or incomplete, then upgrade in-session only to verified Git scope for the same Base User ID and canonical workspace root. Discard pending fallback writeback during that upgrade. Missing, unreadable, malformed, or conflicting scope authority outside this narrow exception must not silently fall back or rebind.
  • Repository resolution is read-only and must not execute Git merely to derive memory identity. Preserve path canonicalization, Git marker validation, symlink/junction containment, and fail-closed behavior.
  • Consume turn metadata only after the matching downstream operation is accepted. Rejection, missing content, interruption, or concurrent replacement must retain or discard metadata with an explicit reason.
  • Backend connection, token, PID, Hook generation, and lifecycle authority use versioned private records. Cross-process read/modify/write requires bounded locking or equivalent version validation; in-memory serialization is not sufficient.
  • Keep the default Backend on loopback. External binding requires explicit opt-in and authentication.

Read the full file on GitHub · 141 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. today Changed · +9 lines · +138 tokens per session db74ac93c306
  2. 4d ago First seen · 132 lines · 1,519 tokens per session scan A 21e793e2f5ae

Subscribe to this mod's changes

memorax-code AGENTS.md is an instructions file published in the GitHub repository memorax-ai/memorax-code (1,196 stars, last pushed today), licensed MIT. It adds 1,657 tokens to every session, about $0.0083 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.

Related

Other instructions, from other repositories

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 tokens

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,182 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,345 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

next.js AGENTS.md

Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens