deepclause-sdk: Instructions file for Codex

AGENTS.md

deepclause-sdk AGENTS.md is an instructions file for Codex, OpenCode from deepclause/deepclause-sdk. It costs 1,651 tokens per session, scanned A, original, MIT.

Repository instructions for deepclause-sdk, covering how AI agents use its command-line tool, compile programs, execute them, and run one-shot tasks.

In plain words
What is it for?
Use them when developing or reviewing changes in the deepclause-sdk repository.
Why use it?
They provide project-specific guidance for working with the SDK, but the description does not give more detail about individual rules.

Instructions file for CodexOpenCode

Written for Codex and OpenCode: the file is AGENTS.md.

This is deepclause/deepclause-sdk's own configuration. It tells Codex and OpenCode how to work on deepclause-sdk 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 deepclause-sdk configures →

Reuse

Borrowing it

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

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 deepclause-sdk AGENTS.md

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

Your own site · 80×15
<a href="https://agentmods.dev/instructions/deepclause/deepclause-sdk/agents-md"><img src="https://agentmods.dev/badge/instructions/deepclause/deepclause-sdk/agents-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 1,651 This file is loaded in full into every session.
When invoked 1,651 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 $0.01651 $0.01651
Opus 5.5 $0.00660 $0.00660
Sonnet 5.5 $0.00330 $0.00330
Haiku 4.5 $0.00165 $0.00165

Measured 8d ago against content hash 92f7f5a95a8e, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

deepclause-sdk 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 · 148 lines

How it starts

The opening of the file, as written. The whole thing — 148 lines — stays where its author put it; the contents beside it link to each section on GitHub.

DeepClause: Guide for AI Agents

DeepClause is a neuro-symbolic runtime for self-modifying agent harnesses. A harness is a living DML program — Prolog with model calls, semantic judgments, and constraints. You (an LLM agent) can read it, extend it, test it, and run it from an interactive REPL; the run command executes a harness one-shot; the TypeScript SDK embeds the runtime.

There is no Markdown→DML compiler. DML is authored directly (by you), and a harness is a normal, inspectable Prolog module.

Three surfaces: repl, run, SDK.


1. The CLI

deepclause init                 # write .deepclause/config.json
deepclause login                # sign in via pi, pick a default model
deepclause repl                 # interactive living-module REPL
deepclause repl --session last  # resume recorded purpose/harness
deepclause repl --mode work --harness example  # use a declared Work shell
deepclause run skill.dml "arg"  # run one entry of a DML file / .zip bundle
deepclause visualize s.dml -g router/2 --presentation -o router.png

run opens the file as a library, calls agent_main/N (arity = number of positional args) or an explicit --goal, and streams events. visualize draws a skill (--presentation or --spec) as a Mermaid diagram from its listing and docs. Other commands: set-model, show-model, list-models, set-backend, logout.


2. The REPL: a living harness

Two input modes:

  • Plain English (default) — a meta task writes/test DML and commits it (replace-by-predicate). Use this to add skills, add tools, change logic, or ask about the harness.
  • goal: <dml>. — query the current harness directly with Prolog/DML.

Every prompt turn is an edit transaction: the module is snapshotted, the meta task edits a scratch copy and tests it with execute, and the result is committed only if it succeeds. Failed turns leave the harness unchanged.

Sessions, turns, traces, and versions are recorded under .deepclause/repl/ and exposed to DML as harness_* facts (harness_session/5, harness_turn/5, harness_change/5, harness_changed/4, harness_head/4, harness_trace/3). rollback. restores the previous version.

Read the full file on GitHub · 148 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 · +9 lines · +112 tokens per session 92f7f5a95a8e
  2. 28d ago Changed · +52 lines · +773 tokens per session 32b9055334fa
  3. 2mo ago First seen · 87 lines · 766 tokens per session scan A 73152e499201

Subscribe to this mod's changes

deepclause-sdk AGENTS.md is an instructions file published in the GitHub repository deepclause/deepclause-sdk (80 stars, last pushed 8d ago), licensed MIT. It adds 1,651 tokens to every session, about $0.0066 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.

Related

deepclause-sdk AGENTS.md compared with instructions from other repositories

deepclause-sdk AGENTS.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
deepclause-sdk AGENTS.mddeepclause/deepclause-sdkInstructions file808d agoMITA1,651Codex, 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