agentmods.dev

soma: Instructions file for Claude Code

AGENTS.md

soma AGENTS.md is an instructions file for Claude Code, Codex, OpenCode from the-metafactory/soma. It costs 1,720 tokens per session, scanned A, original, MIT.

A repository instruction file explains how to work on Soma, a portable personal AI assistant core that can run through several coding-agent hosts.

In plain words
What is it for?
It is for guiding changes to Soma and verifying them with its Bun tests and type checks.
Why use it?
It gives contributors one set of rules for architecture, supported integrations, scripts, and checks.

Instructions file for Claude CodeCodexOpenCode

Written for Claude Code and Codex and OpenCode: PreToolUse hook event, but also the file is AGENTS.md. Also seen: reads .claude/ paths; mentions CLAUDE.md; mentions subagents.

This is the-metafactory/soma's own configuration. It tells Claude Code, Codex and OpenCode how to work on soma itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything soma configures →

not rated 8repo changed 8d ago A scan 1,720 tokens original MIT
View source ↗ the-metafactory/soma
Reuse

Borrowing it

Nothing to install: this file belongs to the-metafactory/soma. 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/the-metafactory/soma/main/AGENTS.md
Clone the repo
git clone --depth 1 https://github.com/the-metafactory/soma

Made for: Claude Code, 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 soma AGENTS.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/the-metafactory/soma/agents-md/github.svg)](https://agentmods.dev/instructions/the-metafactory/soma/agents-md)
Your own site
<a href="https://agentmods.dev/instructions/the-metafactory/soma/agents-md"><img src="https://agentmods.dev/badge/instructions/the-metafactory/soma/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.

agentmods 80×15 button for soma AGENTS.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/the-metafactory/soma/agents-md"><img src="https://agentmods.dev/badge/instructions/the-metafactory/soma/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 1,720 This file is loaded in full into every session.
When invoked 1,720 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.01720 $0.01720
Opus 5.5 $0.00688 $0.00688
Sonnet 5.5 $0.00344 $0.00344
Haiku 4.5 $0.00172 $0.00172

Measured 8d ago against content hash d6f868bb0f23, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

soma 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 8d 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 · 117 lines

How it starts

The opening of the file, as written. The whole thing — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Soma Agent Instructions

Soma is the substrate-portable Personal AI Assistant core. It extracts the assistant body from any one host: identity, telos, ISA, skills, memory, policy, learning, and adapter contracts.

How To Work In This Repo

  • Treat ISA.md as the current source of truth for scope, criteria, decisions, and verification.
  • Read README.md, docs/architecture.md, and docs/substrate-adapters.md before changing architecture.
  • Keep substrate-specific behavior behind adapter boundaries.
  • Keep the core filesystem-native and model-provider-neutral.
  • Prefer deterministic TypeScript contracts over prompt-only conventions.
  • Use Bun for scripts and tests.
  • To add a top-level command, add a loader to COMMANDS in src/cli.ts and its usage to loadCommandHelp. The loader imports the command's module when the command runs. Hooks call soma policy guard on every tool call and algorithm classify on every prompt, and each module that soma loads up front adds to that start time.
  • Keep the hook hot paths light: src/cli/policy.ts, src/cli/algorithm-classify.ts, and the workers that load them (src/cli/policy-worker.ts, src/cli/mode-classifier-worker.ts). Inside a component, import from the defining module. Across components, bun run check:boundaries requires the public entry, and some of those are large: src/flows/index.ts pulls in about 3 MB. Move an import that only some actions use inside those actions, as src/cli/algorithm.ts does for ../flows/index.
  • Verify changes with bun run test and bun run typecheck (bun run test passes --timeout=60000, since bun's default 5 s per test is too short for the install tests on a loaded host, and --parallel=6, which runs the files across six workers; one worker per core timed out install tests on a loaded laptop).

Enforced rules

Each rule pairs with the check that enforces it and the commit it was proven on: the check fails on that commit's mistake and passes on the fix. When a rule here is broken and nothing failed, fix the check, not only the code.

Read the full file on GitHub · 117 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. 8d ago Changed · +13 lines · +258 tokens per session d6f868bb0f23
  2. 19d ago Changed · +10 tokens per session e72b9160bcb4
  3. 21d ago Changed · +13 lines · +635 tokens per session 982b8d9e5c7b
  4. 28d ago Changed · +34 tokens per session 29f0df9163f6
  5. 2mo ago First seen · 91 lines · 783 tokens per session scan A 846fa49aaef8

Subscribe to this mod's changes

soma AGENTS.md is an instructions file published in the GitHub repository the-metafactory/soma (8 stars, last pushed yesterday), licensed MIT. It adds 1,720 tokens to every session, about $0.0069 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 2026-08-31.

Related

soma AGENTS.md compared with instructions from other repositories

soma AGENTS.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
soma AGENTS.mdthe-metafactory/somaInstructions file8yesterdayMITA1,720Claude Code, Codex, OpenCode
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143kyesterdayMITA7,461Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchainInstructions file149kyesterdayMITA4,680Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA5,516GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA3,093GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harnessInstructions file274k2d agoMITA4,313Codex, OpenCode