What is mnemonica AGENTS.md?

Repository instructions for agents modifying the mnemonica library. They require the agent to pause and ask the user when errors, uncertainty, assumptions, or certain file-creation situations arise.

Use them before making library changes, before delegating work to multiple agents, and whenever a tool error, unclear intent, possible assumption, or framework-specific rule appears.

They prevent agents from guessing or continuing after a problem when the project requires explicit user direction.

How wentout/mnemonica configures Codex and OpenCode

Reuse

Borrowing it

Nothing to install: this file belongs to wentout/mnemonica. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/wentout/mnemonica/master/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/wentout/mnemonica

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.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/wentout/mnemonica/agents-md/github.svg)](https://agentmods.dev/instructions/wentout/mnemonica/agents-md)

Your own site

<a href="https://agentmods.dev/instructions/wentout/mnemonica/agents-md"><img src="https://agentmods.dev/badge/instructions/wentout/mnemonica/agents-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/wentout/mnemonica/agents-md"><img src="https://agentmods.dev/badge/instructions/wentout/mnemonica/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 5,386 This file is loaded in full into every session.

When invoked 5,386 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

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.

Model	Per session	Once invoked
Fable 5.1	$0.05386	$0.05386
Opus 5.5	$0.02154	$0.02154
Sonnet 5.5	$0.01077	$0.01077
Haiku 4.5	$0.00539	$0.00539

Measured 1mo ago against content hash fb1ebc6cfdce, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

mnemonica 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 1mo 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 · 447 lines

How it starts

The opening of the file, as written. The whole thing — 447 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md

Three standing rules, all non-negotiable:

Ask before doing. When anything is unclear, surprising, or irreversible — stop and ask first. One question is cheaper than a wrong change.
A question is not a task. Answer from what you know, zero tool calls — investigation is for requested changes, never for answers.
Be lazy. A lazy developer never builds what needs re-doing: no half-work, no throwaway scaffolding, no "good enough for now" that a later session must unpick. Do it once, complete — or don't start.

This file provides guidance specific to mnemonica/core for AI agents modifying the library itself. If you are using mnemonica in your own project, start with README.md.

Before any subagent fan-out or swarm: read .ai/SWARM.md. Stop-and-ask is mandatory there, exactly as in Rule #1 below. A subagent's report is a claim, not evidence. Before relaying any of it, audit the actual code it changed — read the diff, understand every hunk, re-run its "green" claims live. If you cannot explain a change line by line, you have not reviewed it. Delegation never transfers responsibility.

Rule #1 — Non-negotiable: PAUSE AND ASK

This is the highest-priority rule. It overrides everything else in this file.

STOP and ask the user before proceeding when ANY of these is true:

An error occurred — Edit failed, Write produced unexpected results, or any tool returned an error. A successful retry does NOT cancel this rule: report the error AND how you recovered.
You are uncertain — about what change to make, how a function works, or what the user intended.
You are filling gaps with assumption — "probably", "likely", "I think", "it should work" are signals to stop.
A tool or environment constraint conflicts with the user's stated preference — surface the conflict verbatim; never silently satisfy the tool at the user's expense.
You are about to create or duplicate a file the user did not explicitly request — ask first. Never copy content when mv, rename, or a reference suffices.

Read the full file on GitHub · 447 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.

1mo ago Changed · +12 lines · +215 tokens per session fb1ebc6cfdce
1mo ago Changed · -2 lines · -35 tokens per session 4264728b9200
1mo ago Changed · +22 lines · +238 tokens per session e48390d69a24
1mo ago Changed · +2 lines · +68 tokens per session 4f01364e91d5
1mo ago Changed · +63 lines · +724 tokens per session 82cb25d60001
2mo ago Changed · +7 lines · +128 tokens per session a5786b0bc51a
2mo ago First seen · 343 lines · 4,048 tokens per session scan A 9d7c62450c66

Subscribe to this mod's changes

mnemonica AGENTS.md is an instructions file published in the GitHub repository wentout/mnemonica (41 stars, last pushed yesterday), licensed MIT. It adds 5,386 tokens to every session, about $0.0215 per session on Opus 5.5. 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.

mnemonica AGENTS.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
mnemonica AGENTS.mdwentout/mnemonica	Instructions file	41	yesterday	MIT	A	5,386	Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	today	MIT	A	7,461	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	5,516	GitHub Copilot
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	today	MIT	A	4,680	Codex, OpenCode
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	today	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	275k	3d ago	MIT	A	4,313	Codex, OpenCode

mnemonica: Instructions file for Codex