bento AGENTS.md

bento AGENTS.md is an instructions file for Codex, OpenCode from kajogo777/bento. It costs 4,129 tokens per session, scanned C, original, Apache-2.0.

Repository instructions for Bento, a Go tool that saves an AI agent's workspace state—such as dependencies, caches, memory, and session data—as portable OCI artifacts. OCI artifacts are standard packages stored locally or in a container registry.

In plain words
What is it for?
Use them when working on Bento's command-line interface, workspace packaging, registry storage, extensions, secret handling, configuration, or lifecycle hooks.
Why use it?
They explain how the project is organized and identify the design decisions that contributors and coding agents must preserve.

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/kajogo777/bento/agents-md
Clone the repo
git clone --depth 1 https://github.com/kajogo777/bento

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 bento AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/kajogo777/bento/agents-md.svg)](https://agentmods.dev/instructions/kajogo777/bento/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/kajogo777/bento/agents-md"><img src="https://agentmods.dev/badge/instructions/kajogo777/bento/agents-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 4,129 This file is loaded in full into every session.
When invoked 4,129 The same file — it is already loaded in full.
Security scan C 1 finding. 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 5.1 $0.04129 $0.04129
Opus 5 $0.02065 $0.02065
Sonnet 5 $0.00826 $0.00826
Haiku 4.5 $0.00413 $0.00413

Measured 5d ago against content hash 82bf331f9bdf, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade C, and why

bento AGENTS.md scanned grade C with 1 finding 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 5d 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.

Recursive force deletehighDestructive command

rm -rf with a variable or a broad path is one typo away from removing the wrong tree.

make clean # rm -rf bin/ dist/
AGENTS.md · 314 lines

How it starts

The opening of the file, as written. The whole thing — 314 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS.md - Bento Development Guide

Project Overview

Bento packages AI agent workspace state into portable, layered OCI artifacts. It captures everything git doesn't: agent memory, installed dependencies, build caches, conversation history, and session state. Checkpoints are stored as standard OCI images in any container registry.

Repository: github.com/kajogo777/bento Language: Go (1.25+) License: Apache 2.0

Architecture

cmd/bento/main.go           → entrypoint, version injection
internal/
  cli/                       → cobra commands (init, save, open, list, diff, etc.)
  config/                    → bento.yaml parsing, validation, platform defaults
  extension/                 → composable extensions (agent, deps, tool detection)
  workspace/                 → file scanning, layer packing (tar+gzip), .bentoignore
  registry/                  → OCI store (local image layout + remote push)
  manifest/                  → OCI manifest/config construction, annotations, DAG
  secrets/                   → secret scanning (gitleaks), env hydration, providers, scrubbing, backends
  hooks/                     → lifecycle hook execution (pre_save, post_restore, etc.)
  policy/                    → garbage collection (retention tiers, blob pruning)
  watcher/                   → file-system watcher for auto-checkpointing

Key Design Decisions

  1. Standard OCI media types - All layers use application/vnd.oci.image.layer.v1.tar+gzip so docker pull, COPY --from, and containerd work natively. Layer semantics are carried by org.opencontainers.image.title annotations.

  2. Shared blob store - Local OCI store at ~/.bento/store/ uses a shared content-addressed blob pool. Identical layers across workspaces are stored once.

  3. Composable extensions - Each agent framework, language, and tool gets a small extension that contributes patterns to the right layer. Extensions auto-detect and merge. No monolithic harnesses.

  4. Three core layers - deps (rarely changes, large), agent (changes often, small), project (catch-all). Extensions can add more layers (e.g., build-cache).

Read the full file on GitHub · 314 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. 5d ago First seen · 314 lines · 4,129 tokens per session scan C 82bf331f9bdf

Subscribe to this mod's changes

bento AGENTS.md is an instructions file published in the GitHub repository kajogo777/bento (22 stars, last pushed 4mo ago), licensed Apache-2.0. It adds 4,129 tokens to every session, about $0.0206 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.