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/dosco/graphjin/agents-mdgit clone --depth 1 https://github.com/dosco/graphjinWhat 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 | $0.05837 | $0.05837 |
| Opus 5 | $0.02919 | $0.02919 |
| Sonnet 5 | $0.01167 | $0.01167 |
| Haiku 4.5 | $0.00584 | $0.00584 |
Grade A, and why
graphjin 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 yesterday.
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 — 396 lines — stays where its author put it; the contents beside it link to each section on GitHub.
GraphJin Agent Guide
This document is a guide for AI agents working on the GraphJin codebase. It outlines the architectural patterns, coding conventions, and common tasks associated with maintaining and extending this library.
Architectural Overview
GraphJin is a compiler that turns GraphQL into database queries. For SQL databases, it generates SQL. For MongoDB, it generates a JSON DSL that is translated to aggregation pipelines. It is NOT a typical ORM or resolver-based GraphQL server.
- Core Philosophy: Push as much work as possible to the database.
- No Resolvers: Data fetching is done via a single generated SQL query. Do not add resolvers for database fields.
- Schema Driven: The database schema (
sdata) is the source of truth.
Directory Structure & Responsibilities
| Path | Component | Responsibility |
|---|---|---|
core/api.go |
Public API | The only entry point for users. Changes here are breaking. |
core/core.go |
Engine | Internal orchestration, initialization, and state management. |
core/internal/sdata |
Schema | Metadata about tables, columns, and relationships. Graph traversal logic. |
core/internal/qcode |
IR Compiler | Front-end compiler. Parses GraphQL -> QCode (Intermediate Representation). |
core/internal/psql |
SQL Compiler | Back-end compiler. QCode -> SQL. Handles dialect differences. |
core/internal/dialect |
Dialect Interface | Database-specific SQL generation methods. Each dialect implements this interface. |
core/internal/graph |
GraphQL Parser | Lexer and parser for GraphQL input. Builds AST. |
core/internal/jsn |
JSON Processing | High-performance JSON parsing and filtering utilities. |
serv/ |
HTTP Service | Standalone server with REST, GraphQL, and WebSocket APIs. |
auth/ |
Authentication | Auth providers: JWT, Auth0, Firebase, Rails session. |
cmd/ |
CLI Tool | Command-line interface for migrations, deployment, and management. |
conf/ |
Configuration | YAML-based config loading and validation. |
wasm/ |
WebAssembly | WASM build for NodeJS integration. |
mongodriver/ |
MongoDB Driver | Custom database/sql-compatible driver for MongoDB. Translates JSON DSL to aggregation pipelines. |
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.
- yesterday First seen · 396 lines · 5,837 tokens per session scan A ee1c41b86ca1
graphjin AGENTS.md is an instructions file published in the GitHub repository dosco/graphjin (3,153 stars, last pushed yesterday), licensed Apache-2.0. It adds 5,837 tokens to every session, about $0.0292 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.
Other instructions, from other repositories
whodb AGENTS.md
Instructions for clidey/whodb, covering whodb development guide, git commits — hard rule, terminology, non-negotiable rules and behavioral guidelines.
whodb CLAUDE.md
Instructions for clidey/whodb, a project described as: Where data access meets operational intelligence.
Chat2DB AGENTS.md
AGENTS.md instructions for OtterMind/Chat2DB, covering chat2db community agent contract, context, product invariants, repository map and structural code navigation.
leoric AGENTS.md
Instructions for cyjake/leoric, covering agent guidelines, project overview, commands, architecture and conventions.
postgrebase AGENTS.md
Instructions for zhenruyan/postgrebase, covering postgrebase — agent development guide, tech stack, project structure, build & run and build (static, no cgo).
hs-sql-agent AGENTS.md
Instructions for tse-wei-chen/hs-sql-agent, covering agents.md, repo structure, commands, backend and run locally.