prime-silo AGENTS.md

Repository-wide rules for prime-silo, with documentation treated as part of how the software works. It defines which AGENTS.md file owns each area and requires documentation updates alongside relevant code changes.

In plain words
What is it for?
Use it when changing repository structure, application, server, command, packaging, or related documentation.
Why use it?
It helps agents make changes in the correct layer and prevents architecture or workflow knowledge from becoming outdated.

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/binary16labs/prime-silo/agents-md
Clone the repo
git clone --depth 1 https://github.com/binary16labs/prime-silo

Made for: Codex, OpenCode.

Per session 6,529 This file is loaded in full into every session.
When invoked 6,529 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 $0.06529 $0.06529
Opus 5 $0.03265 $0.03265
Sonnet 5 $0.01306 $0.01306
Haiku 4.5 $0.00653 $0.00653

Measured 3d ago against content hash 8288ae06939f, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

prime-silo 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 · 313 lines

How it starts

The opening of the file, as written. The whole thing — 313 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AGENTS

Documentation First

Documentation is the most important part of this project.

Treat every AGENTS.md file as part of the runtime contract, not as optional notes. Poor documentation causes agent behavior drift, architecture drift, and bad changes in the wrong layer.

This repository now uses a documentation hierarchy:

  • /AGENTS.md owns repo-wide rules, documentation policy, and top-level architecture
  • the five core docs are /AGENTS.md, /app/AGENTS.md, /server/AGENTS.md, /commands/AGENTS.md, and /packaging/AGENTS.md
  • deeper AGENTS.md files inside app/ and server/ own the concrete implementation contracts for the module or subsystem in their subtree
  • the closer the doc is to the code, the more technical and specific it should be
  • the higher the doc is in the tree, the more it should focus on principles, ownership, stable contracts, and architecture

Always update the relevant docs in the same session as the code change:

  • update the closest owning AGENTS.md for the files you changed
  • reflect stable architecture, workflow, and agent-facing behavior changes into the supplemental docs under app/L0/_all/mod/_core/documentation/docs/
  • update parent docs too when the higher-level contract, ownership boundary, architecture, or workflow changed
  • keep higher-level docs abstract where appropriate and push implementation detail down into local docs
  • keep lower-level docs concrete, explicit, and practical
  • remove stale or contradictory documentation immediately
  • keep README.md as the public product source of truth for the project pitch, quick starts, community links, release links, and the high-level documentation map
  • do not let README.md become a competing implementation contract; durable architecture and workflow contracts belong in AGENTS.md, while broader agent-facing narrative docs belong in app/L0/_all/mod/_core/documentation/docs/

Documentation depth model:

  • level 0 repo doc: mission, top-level architecture, cross-cutting rules, and the documentation policy for the whole tree
  • level 1 core domain docs: app/, server/, commands/, and packaging/ architecture, plus the template for their immediate child docs
  • level 2 subsystem or module docs: one major area, its owned files, its stable contracts, and the template for any deeper docs in that subtree
  • level 3 leaf docs: one concrete surface, feature, view, or service contract with exact implementation guidance
  • if a level 2 or level 3 doc later gains children, it stops being a leaf and must add its own Documentation Hierarchy section before those child docs land

Read the full file on GitHub · 313 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 · 313 lines · 6,529 tokens per session scan A 8288ae06939f

Subscribe to this mod's changes

prime-silo AGENTS.md is an instructions file published in the GitHub repository binary16labs/prime-silo (5 stars, last pushed 10d ago), licensed MIT. It adds 6,529 tokens to every session, about $0.0326 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 2026-08-31.