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/aaronsb/knowledge-graph-system/claude-mdgit clone --depth 1 https://github.com/aaronsb/knowledge-graph-systemWhat 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 | $0.02036 | $0.02036 |
| Opus 5 | $0.01018 | $0.01018 |
| Sonnet 5 | $0.00407 | $0.00407 |
| Haiku 4.5 | $0.00204 | $0.00204 |
Grade D, and why
knowledge-graph-system CLAUDE.md scanned grade D with 3 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
sudo ./operator.sh upgrade # Pull images, migrate, restart apps Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
curl -fsSL https://raw.githubusercontent.com/aaronsb/knowledge-graph-system/main/install.sh | bash Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -fsSL https://raw.githubusercontent.com/aaronsb/knowledge-graph-system/main/install.sh | bash How it starts
The opening of the file, as written. The whole thing — 259 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Knowledge Graph System - Claude Development Guide
Project Overview
A multi-dimensional knowledge extraction system that transforms documents into interconnected concept graphs using Apache AGE (PostgreSQL graph extension).
Fully Containerized - no local Python installation required.
┌─────────────────────────────────────────────────────────────┐
│ Docker Containers │
├─────────────────────────────────────────────────────────────┤
│ Documents → [API Container] → LLM Extraction → [Postgres] │
│ ↓ ↓ │
│ [Garage S3 Storage] [Apache AGE Graph] │
│ ↓ ↓ │
│ [Web Container] ← REST API ← [API Container] │
│ │
│ Configuration: [Operator Container] → [Postgres Config] │
└─────────────────────────────────────────────────────────────┘
↑
kg CLI (optional)
MCP Server (optional)
Tech Stack:
- Python 3.11+ FastAPI (REST API server)
- Apache AGE / PostgreSQL 16 (graph database, openCypher)
- TypeScript/Node.js (CLI + MCP)
- React/Next.js (web visualization)
- Garage (S3-compatible storage)
- Docker Compose (orchestration)
Quick Start
Production deployment (standalone installer):
curl -fsSL https://raw.githubusercontent.com/aaronsb/knowledge-graph-system/main/install.sh | bash
Development setup (from repo):
git clone https://github.com/aaronsb/knowledge-graph-system.git
cd knowledge-graph-system
./operator.sh init # Guided setup
./operator.sh start # Daily start
Optional CLI:
cd cli && npm run build && ./install.sh
kg health
Platform Lifecycle
operator.sh is the primary management tool (apt-style commands):
# Updates (pull images, no restart)
./operator.sh update # Pull all images
./operator.sh update api # Pull specific service
# Upgrades (pull, migrate, restart)
./operator.sh upgrade # Full upgrade
./operator.sh upgrade --dry-run # Preview changes
# Daily operations
./operator.sh start # Start platform
./operator.sh stop # Stop platform
./operator.sh status # Check health
./operator.sh logs api -f # Follow API logs
./operator.sh shell # Configuration shell
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 · 259 lines · 2,036 tokens per session scan D b306ab82413f
knowledge-graph-system CLAUDE.md is an instructions file published in the GitHub repository aaronsb/knowledge-graph-system (124 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 2,036 tokens to every session, about $0.0102 per session on Opus 5. A static security scan graded it D with 3 findings (asks for root, downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
kartograph CLAUDE.md
Instructions for openshift-hyperfleet/kartograph, a project described as: Enterprise-Ready Bi-Temporal Knowledge Graph Platform.
tiding CLAUDE.md
Instructions for iXanadu/tiding, a project described as: Tiding (formerly engram) — durable shared semantic memory + inter-agent messaging for AI agents. FastAPI + pgvector, hybrid semantic/trigram search, durable inbox with presence — single-operator by design.
deployplan CLAUDE.md
Claude Code instructions for jsuncar/deployplan, a project described as: Deploy Next.js without Vercel. Supabase-like Postgres + public URL. Agent MCP: npx -y deployplan init.
YTSearch timestamp.instructions.md
Timestamp management standards and best practices for database models and service layers.
m_flow AGENTS.md
AGENTS.md instructions for FlowElement-xinliuyuansu/m_flow, covering m-flow — developer & agent reference, 1. repository map, extension points, 2. local development and python backend (requires python 3.10 – 3.13).
kglite CLAUDE.md
Instructions for kkollsga/kglite, covering kglite — claude code conventions, build & test, architecture, the boundary principle (wrappers vs core) — summary and in-memory is the core product.