What is codex-agent-mem AGENTS.md?

Repository instructions for codex-agent-mem, a local memory layer that helps MCP-compatible coding agents retain information between sessions.

Use them when modifying the memory system, its SQLite database, MCP interface, FastAPI screen, or generated AGENTS.md context.

They explain the project’s names, storage, retrieval interfaces, memory extraction, and maintenance rules.

How MarceloCaporale/codex-agent-mem configures Codex and OpenCode

Reuse

Borrowing it

Nothing to install: this file belongs to MarceloCaporale/codex-agent-mem. 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/MarceloCaporale/codex-agent-mem/main/AGENTS.md

Clone the repo

git clone --depth 1 https://github.com/MarceloCaporale/codex-agent-mem

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/marcelocaporale/codex-agent-mem/agents-md/github.svg)](https://agentmods.dev/instructions/marcelocaporale/codex-agent-mem/agents-md)

Your own site

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

Per session 1,043 This file is loaded in full into every session.

When invoked 1,043 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	$0.01043	$0.01043
Opus 5.5	$0.00417	$0.00417
Sonnet 5.5	$0.00209	$0.00209
Haiku 4.5	$0.00104	$0.00104

Measured 2mo ago against content hash 6b000860e428, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

codex-agent-mem 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 2mo 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 · 116 lines

How it starts

The opening of the file, as written. The whole thing — 116 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS

This repository is intentionally optimized for MCP-compatible AI agents, coding workflows, deep-research tools, and maintainers who need a fast map of what exists today.

Public name vs import name

Public repository name: codex-agent-mem
Python package import name: codex_agent_mem
CLI commands use the public name with hyphens, for example codex-agent-mem-smoke

What this project is

A portable, local-first MCP memory layer for agent continuity across MCP-compatible runtimes
Current durable store: local SQLite
Current retrieval surface: MCP stdio
Current capture path: local agent turn ingestion, including Codex notify on agent-turn-complete
Current inspection surface: local FastAPI UI at /ui
Current continuity surface: generated AGENTS.md working-memory block when compression is favorable
Current extraction strategy: heuristic extraction of session summaries, user requests, objectives, decisions, constraints, pending/completed work, blockers, DoD items, and completion claims

Working rules for agents

Preserve the local-first MCP core.
Do not use an operational maintainer database for tests; use temporary databases.
Keep hosted/web bridge experiments out of the public core unless the repository scope changes explicitly.
Keep compatibility and token-savings claims tied to reproducible evidence.
Keep the six README files equivalent in content across languages.
For broad/container project keys, prefer mem_bootstrap_context or mem_scope_resolve before loading mem_context_pack(project_key). A project-wide pack may remain available, but it must not be treated as the active thread when candidate lanes are ambiguous.
Do not tag, publish, or release from a dirty tree.

Fastest commands

pip install -e .[dev]
pytest -q
python -m compileall src
ruff check .
python scripts/mcp_contract_smoke.py --both
python scripts/check_repo_hygiene.py
python scripts/smoke_release.py --mcp-subprocess --with-ruff --with-build --with-wheel-smoke
codex-agent-mem-smoke
codex-agent-mem-bootstrap-codex --db-path C:\Users\YOU\.codex_agent_mem\codex_agent_mem.db
python -m build

Read the full file on GitHub · 116 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.

2mo ago First seen · 116 lines · 1,043 tokens per session scan A 6b000860e428

Subscribe to this mod's changes

codex-agent-mem AGENTS.md is an instructions file published in the GitHub repository MarceloCaporale/codex-agent-mem (41 stars, last pushed 6mo ago), licensed Apache-2.0. It adds 1,043 tokens to every session, about $0.0042 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.

Memory and context codex codex-cli local-first mcp sqlite

codex-agent-mem AGENTS.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
codex-agent-mem AGENTS.mdMarceloCaporale/codex-agent-mem	Instructions file	41	6mo ago	Apache-2.0	A	1,043	Codex, OpenCode
repobrain copilot-instructions.mdstudy8677/repobrain	Instructions file	1.4k	14d ago	MIT	A	229	GitHub Copilot
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	today	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot

codex-agent-mem: Instructions file for Codex