opencode-rules AGENTS.md

Repository instructions for developing and releasing the frap129/opencode-rules project. They describe its commands, structure, plugin entry points, and how it injects rules into agent messages.

In plain words
What is it for?
Use them when installing dependencies, linting, type-checking, running tests, understanding plugin exports, or modifying rule injection behavior.
Why use it?
They give an agent the project-specific setup and checks needed to change the repository without overlooking its architecture or conventions.

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/frap129/opencode-rules/agents-md
Clone the repo
git clone --depth 1 https://github.com/frap129/opencode-rules

Made for: Codex, OpenCode.

Per session 1,099 This file is loaded in full into every session.
When invoked 1,099 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 $0.01099 $0.01099
Opus 5 $0.00549 $0.00549
Sonnet 5 $0.00220 $0.00220
Haiku 4.5 $0.00110 $0.00110

Measured 2d ago against content hash b63a862d2b3e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

opencode-rules 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 2d 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 · 46 lines

How it starts

The opening of the file, as written. The whole thing — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Repo Notes

Commands

  • Use Aube and install with aube install --frozen-lockfile. Run checks in this order: aube run lint -> aubx tsc --noEmit -> aube run test:run.
  • There is no typecheck script; typecheck with aubx tsc --noEmit.
  • Run one colocated Vitest file with aube run test:run src/<name>.test.ts (or a path under tui/). tsconfig.json excludes test/spec files, so tsc does not typecheck them.
  • docs/silent-message-implementation.md describes a superseded design; current delivery is synthetic parts via chat.message.

Architecture

  • One package, two default plugin exports: server entry src/index.ts returns { id, server }; TUI entry tui/index.tsx returns { id, tui }. The package root imports built dist/src/index.js, and "./tui" imports built dist/tui/index.js (its types come from dist/tui/index.d.ts).
  • OpenCode's plugin loader calls every named export as a plugin initializer; any added named export must be callable (see the __testOnly pattern in src/index.ts).
  • Each injection event is one <system-message> block with one preamble and <rule name="..."> blocks using frontmatter name or the filename stem. Session-durable rules (unconditional, globs, fileContains, keywords, command, project, os, ci) are injected as one synthetic text part appended to the user message via chat.message, not into the system prompt. A durable rule first matched by a live File observation is instead admitted at the earliest applicable dispatch through an awaited no-reply session.prompt, with transient fallback and retry when persistence fails. Ephemeral rules (agent, model, branch, tools) are delivered as one transient synthetic message per matching turn via experimental.chat.messages.transform and never persisted. Live tool.execute.after observations are the sole matching source for globs/fileContains; history rebuilds Working-context paths (compaction projection) and delivery-ledger identity only, never File observations. Path-derived identity keys prevent durable rules still present in history from being re-appended after content edits or resume; message removal invalidates the ledger so reverted delivery is retried, and after compaction the ledger is rebuilt so missing durable rules are re-appended. Hook guidance uses the same framing.
  • Depends on experimental OpenCode hooks (experimental.chat.messages.transform, experimental.session.compacting); re-verify against @opencode-ai/plugin when upgrading.
  • Per-session matched-rule state is written atomically to ~/.opencode/state/opencode-rules/{sessionID}.json, which the TUI sidebar reads.

Read the full file on GitHub · 46 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. 2d ago First seen · 46 lines · 1,099 tokens per session scan A b63a862d2b3e

Subscribe to this mod's changes

opencode-rules AGENTS.md is an instructions file published in the GitHub repository frap129/opencode-rules (112 stars, last pushed 2d ago), licensed MIT. It adds 1,099 tokens to every session, about $0.0055 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 2026-08-30.

Related

Other instructions, from other repositories

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,182 tokens

buildNext

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

next.js AGENTS.md

Instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

spec-kit AGENTS.md

Instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,040 tokens

langchain AGENTS.md

Instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,345 tokens