redthread AGENTS.md

AGENTS.md instructions for matheusht/redthread, covering redthread agent architecture, 1. core operating guidelines, 2. the orchestration workflow (principal vs subagents), the principal agent must and subagent usage.

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

Made for: Codex, OpenCode.

Per session 1,929 This file is loaded in full into every session.
When invoked 1,929 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 $0.01929 $0.01929
Opus 5 $0.00964 $0.00964
Sonnet 5 $0.00386 $0.00386
Haiku 4.5 $0.00193 $0.00193

Measured today against content hash aa362dac7083, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

redthread 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 today.

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 · 130 lines

How it starts

The opening of the file, as written. The whole thing — 130 lines — stays where its author put it; the contents beside it link to each section on GitHub.

RedThread Agent Architecture

This document dictates the behavior and orchestration of the Antigravity Agent when working within the RedThread repository.

All behavioral configurations must strictly refer to the docs/ sources of truth.

1. Core Operating Guidelines

  • Decision Matrix: Before running any operation, the agent must consult docs/AGENT_DECISION_TREE.md to identify which domain document to load based on the user's intent.
  • Working Methodology: All tasks must follow the RPI (Research → Plan → Implement) flow outlined in docs/RPI_METHODOLOGY.md. Context must not exceed 40% window utilization.
  • Default Response Style: Use full caveman mode by default unless the user asks for a different style. That means simple words, short direct sentences, practical structure, low fluff, and clear "what this is / why it matters / what next" guidance.

2. The Orchestration Workflow (Principal vs Subagents)

Antigravity operates as the Principal Agent inside the RedThread ecosystem. It acts identically to the LangGraph supervisor documented in docs/PHASE_REGISTRY.md and docs/AGENT_ARCHITECTURE.md—it manages the task graph while delegating execution.

The Principal Agent Must:

  1. Clarify intent.
  2. Load relevant .agent/rules/.
  3. Read the relevant document from docs/AGENT_DECISION_TREE.md.
  4. Trigger the correct .agent/skills/.

Subagent Usage

When tasks bridge boundaries, the Principal Agent MUST delegate:

  • Research Agent (Model: Opus 4.6): Use for sweeping file aggregation or reading large datasets. Focuses strictly on extracting context, paths, and patterns without modifying files.
  • Plan Agent / Implement Agent: Used to isolate complex edits (e.g. creating a PersonaGenerator node) from the Principal Agent's context. Always requires explicit .agent/skills/ procedures like TAP, PAIR, or G-Eval execution.

3. Mandatory Component Rules

Do not maintain rules in this document. Always apply .agent/rules/ for global operations. Use .agent/skills/ for specific tasks.

Read the full file on GitHub · 130 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. today First seen · 130 lines · 1,929 tokens per session scan A aa362dac7083

Subscribe to this mod's changes

redthread AGENTS.md is an instructions file published in the GitHub repository matheusht/redthread (47 stars, last pushed 10d ago), licensed MIT. It adds 1,929 tokens to every session, about $0.0096 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 2026-09-01.