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.
npx agentmods add instructions/guodaxia103/tabrix/agents-mdgit clone --depth 1 https://github.com/guodaxia103/tabrixWhat 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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.05263 | $0.05263 |
| Opus 5 | $0.02632 | $0.02632 |
| Sonnet 5 | $0.01053 | $0.01053 |
| Haiku 4.5 | $0.00526 | $0.00526 |
Grade A, and why
tabrix 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.
How it starts
The opening of the file, as written. The whole thing — 265 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Tabrix AI Development Rules
This repository welcomes contributions from any AI coding assistant — Claude (Desktop / Code / Cursor), Codex CLI, Cursor Agent, Cline, Windsurf, GPT-5, Gemini, and any future tool capable of reading + writing files. The historical Claude → Codex two-name pairing that appears in old sprint retros was about two specific tools the original maintainer happened to be running, not a restriction on who may contribute.
What matters is the role an AI plays for a given task (see "Tiered Execution Model" below), not which vendor ships it.
This repository uses a strict "small, verifiable, low-regression" workflow for AI-assisted development.
All code changes in this repository must follow the karpathy-guidelines skill:
- think before coding
- keep the change as small as possible
- define a verifiable success condition
- prefer a real fix over a temporary suppression
Mandatory Read Path
Before changing code or docs, every AI assistant should read the public repository docs below first, in this order:
- README.md
- docs/README.md
- docs/ROADMAP.md — public product direction and contribution areas.
- docs/PRODUCT_SURFACE_MATRIX.md — current public product surface + capability tiers (GA/Beta/Experimental/Internal).
- docs/ARCHITECTURE.md
- docs/PROJECT_STRUCTURE.md
- CONTRIBUTING.md
Then continue with the smallest relevant task-specific reading set below.
Private maintainer planning boundary
Detailed PRDs, sprint backlogs, owner-lane briefs, and version-internal plans are maintainer-private materials and must stay outside the public Git tree. Public contributors should use docs/ROADMAP.md, docs/PRODUCT_SURFACE_MATRIX.md, and code-local issues/PRs as the public working contract.
Task-Specific Reading
Product positioning, public docs, and boundary-governance tasks
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.
- 2d ago First seen · 265 lines · 5,263 tokens per session scan A bc63f8675648
tabrix AGENTS.md is an instructions file published in the GitHub repository guodaxia103/tabrix (5 stars, last pushed 3mo ago), licensed MIT. It adds 5,263 tokens to every session, about $0.0263 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.
Other instructions, from other repositories
openbrowser CLAUDE.md
Claude Code instructions for dylansantwani/openbrowser, covering openbrowser — working guidance, what this is, hard rules, where the token budget goes and things that are true and surprising.
nekoro-browser AGENTS.md
Instructions for zeshuochen/nekoro-browser, covering code priorities, testing and contributing.
openbrowser AGENTS.md
AGENTS.md instructions for dylansantwani/openbrowser, covering agents.md, what this is, build, test, run, hard rules and the token budget is a first-class concern.
AIPex AGENTS.md
Instructions for AIPexStudio/AIPex, covering package architecture, building and running, writing tests, test structure and framework and mocking (vi from vitest).
opentabs AGENTS.md
Instructions for opentabs-dev/opentabs, covering project instructions for codex, project overview, architecture, you are connected to this project's mcp server and workflow skills.
browser-bridge AGENTS.md
Instructions for koltyakov/browser-bridge, covering project guide, data flow, package map, where to find things and method dispatch (extension side).