agentmods.dev

Wax AGENTS.md

A set of instructions for AI coding agents working on Wax, including rules for public files, repository navigation, and changes to its public API. Public repository hygiene means keeping private notes, temporary files, and internal planning material out of a public codebase.

In plain words
What is it for?
Use it when editing Wax source code, tests, documentation, examples, release files, or public integrations.
Why use it?
It helps prevent accidental publication of private or irrelevant material and keeps changes reviewable.

Instructions file for CodexOpenCode

790 4d ago A scan 1,403 tokens original Apache-2.0
View the source file christopherkarani/Wax
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/christopherkarani/wax/agents-md
Clone the repo
git clone --depth 1 https://github.com/christopherkarani/Wax

Made for: Codex, OpenCode.

Per session 1,403 This file is loaded in full into every session.
When invoked 1,403 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 $0.01403 $0.01403
Opus 5 $0.00701 $0.00701
Sonnet 5 $0.00281 $0.00281
Haiku 4.5 $0.00140 $0.00140

Measured 3d ago against content hash 33688989127f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

Wax 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 · 82 lines

How it starts

The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Wax Agent Instructions

Public Repository Hygiene

This repository is the public, front-facing Wax source tree. Keep it limited to source code, tests, release automation, product documentation, examples, and public integration assets.

Do not commit or track:

  • planning scratchpads, execution ledgers, remediation checklists, or task logs
  • agent lessons, session notes, handoff notes, or private workflow rules
  • marketing drafts for articles, social posts, launch copy, or prompt ideation
  • screenshots, issue snapshots, browser captures, or temporary debugging images
  • loose root-level analysis reports that are not part of the official docs
  • customer, investor, pricing, launch, or unreleased product planning material
  • .grok/, Resources/npm/waxmcp/dist/, __pycache__/, *.pyc

Use external/private notes for working plans. If a public plan is genuinely part of the product documentation, place it under an official docs surface and make sure it is written as user-facing documentation, not as an internal task list.

Before committing, run:

git ls-files | rg '(^tasks/|^marketing/|issue[0-9]+_|snapshot|screenshot|TECHNICAL_ANALYSIS|audit-.*ledger|lessons\.md|todo\.md|\.grok/|waxmcp/dist)'

The command should return no matches unless the file is intentionally public and has been reviewed as part of the docs or release surface.

Agents.md and AGENTS.md are the same file on this volume (case-insensitive). Edit one path.

Change Discipline

  • Keep changes scoped to the user request.
  • Preserve unrelated local work and generated artifacts.
  • Prefer tests or concrete verification for source changes.
  • Do not stage build output, caches, screenshots, or local tool directories.
  • Text memory facade: Sources/Wax/Memory.swift. Photo/video public facades: Sources/Wax/PhotoRAG/PhotoMemory.swift, Sources/Wax/VideoRAG/VideoMemory.swift.
  • Public re-exports live only in Sources/Wax/PublicAliases.swift. Do not restore @_exported import.
  • Package-only engines — do not generate app or docs samples against them: MemoryOrchestrator, PhotoRAGOrchestrator, VideoRAGOrchestrator, WaxSession, the Wax actor, MiniLMEmbedder. Use Memory / PhotoMemory / VideoMemory and BuiltInEmbeddings.make.
  • Swift public surface source of truth: Resources/skills/public/wax/references/public-api.md. If constraints.md or the wax skill disagree, follow public-api.md and the Swift public types.
  • Hermes native provider (canonical): Resources/hermes/wax-memory-plugin. npm ship copy: Resources/npm/waxmcp/plugins/hermes. Listed runtime files must stay byte-identical.
  • Operator MCP playbook (using Wax, not changing it): Claude.md and Resources/skills/public/wax-mcp. Swift/framework work: Resources/skills/public/wax.
  • Product examples are allowed once reviewed. Until Examples/ is committed, public paste contract is README + public-api.md + DocsPasteHarnessTests.

Read the full file on GitHub · 82 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 · 82 lines · 1,403 tokens per session scan A 33688989127f

Subscribe to this mod's changes

Wax AGENTS.md is an instructions file published in the GitHub repository christopherkarani/Wax (790 stars, last pushed 4d ago), licensed Apache-2.0. It adds 1,403 tokens to every session, about $0.0070 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.

Building agentsai-agentsdata-sciencemachine-learningmemory-cachememory-hacking
Related

Other instructions, from other repositories

marimo AGENTS.md

Instructions for marimo-team/marimo, covering marimo development guidelines, your primary responsibility is to the project and its users, quick setup, development commands and python.

marimo-team/marimo · 644 tokens

streamlit copilot-instructions.md

Copilot instructions for streamlit/streamlit, covering streamlit repo overview, tech stack, folder structure, shell & build policy and make commands.

streamlit/streamlit · 2,238 tokens

zenml CLAUDE.md

Claude Code instructions for zenml-io/zenml, covering zenml claude code guidelines, project structure, use zenml docs via mcp, code style & quality standards and commenting policy — explain why, not what.

zenml-io/zenml · 4,696 tokens

marimo copilot-instructions.md

Instructions for marimo-team/marimo: For example, if there is a frontend change to make a border thicker, how does this look in run mode, app-view and edit-view? For a backend change, has the user tested different scenarios?

marimo-team/marimo · 91 tokens

zenml AGENTS.md

AGENTS.md instructions for zenml-io/zenml, covering zenml codex agent guidelines, project structure, always-loaded rules, common commands and branches, git, and prs.

zenml-io/zenml · 1,575 tokens

mlops-python-package AGENTS.md

AGENTS.md instructions for fmind/mlops-python-package, covering agents.md, project overview, setup & core commands, definition of done and conventions & idioms.

fmind/mlops-python-package · 1,248 tokens